[00:02:51] Simetrical hooks.txt is the greatest thing since sliced bread [00:02:53] thanks!! [00:04:45] 03(mod) FlaggedRevs blocks/disables Patroller Extension - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16375 (10Wiki.Melancholie) [00:05:07] 03(mod) Add "diff to last sighted version" to watchlist - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16637 (10JSchulz_4587) [00:13:29] 03aaron * r44553 10/trunk/phase3/includes/specials/SpecialRecentchanges.php: Tweak comment [00:18:47] what version of php does mediawiki runs now? [00:18:47] 03aaron * r44554 10/trunk/extensions/FlaggedRevs/specialpages/RevisionReview_body.php: [00:18:47] * This needs 'PatrolLog::record( $rc, false );' now [00:18:48] * Break long line [00:20:45] 03(mod) FlaggedRevs blocks/disables Patroller Extension - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16375 (10JSchulz_4587) [00:39:10] 03(mod) FlaggedRevs blocks/disables Patroller Extension - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16375 (10Wiki.Melancholie) [00:39:33] 03aaron * r44555 10/trunk/extensions/FlaggedRevs/FlaggedArticle.php: Fixed XHTML warning [00:42:29] 03(mod) FlaggedRevs blocks/disables Patroller Extension - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16375 (10Wiki.Melancholie) [01:07:09] 03(mod) problem on Persian title - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16638 (10matthew.britton) [01:30:36] 03(NEW) Code Review buttons can be considered "confusing" - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16639 trivial; normal; MediaWiki extensions: CodeReview; (p858snake) [01:31:04] Need help. anyone around whom have experiance getting the thumbs to work on local installed versions of mediawiki? [01:54:57] yes. you need imagemagick. read the comments in LocalSettings.php and docs. [01:59:44] Kodos2: You running windows or linux? [02:00:22] windows vista [02:00:33] Ok [02:00:59] First of all dont use imagemagick (sorry simonrvm) its as buggy as all hell [02:01:14] Download yourself a copy of Inkscape and install it [02:01:30] http://www.inkscape.org/ [02:03:11] is it easy to install ? I had a really though time trying to figure out how to install imagemagick [02:03:43] Yes, its pretty easy [02:03:46] I can do it :) [02:04:07] (aka it's really easy) [02:04:40] also, imagemagick + windows = no [02:04:51] *Prom_cat agress [02:05:04] ALso the servers are currently timing for some users. [02:05:18] So yeah, if you cant get onto mediawiki wiki thats why [02:07:35] 03(mod) Allow anon users to define which version should be displayed by default - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14321 (10dotkorg) [02:08:46] Kodos2: When you have installed it let me know [02:09:28] great. thanks for the help [02:10:20] Theres more too it than that [02:11:09] I go to the patient page in /config/LocalSettings.php (i just finished config/index.php installation, and it me to go there) and i ge tthis error message http://rafb.net/p/wOBKoF47.html [02:12:11] any ideas? [02:12:25] Octoroks: you need to move LocalSettings out of the /config directory and into the main directory [02:13:13] ok [02:13:19] ok. I finished installing it .. now what ? [02:13:29] http://pastebin.com/f464789e4 [02:13:38] Add those lines to your localsettings.php file [02:14:08] ok.. [02:14:12] brb [02:14:15] k [02:15:57] buggy? WorksForMe [02:16:01] ok.. what now ? [02:16:16] That should be it [02:16:17] i'm not using windoze [02:16:19] Give it a go [02:16:31] simonrvn: Its buggy. [02:18:11] doesn't work [02:19:17] :/ [02:19:27] Error message> [02:19:28] ? [02:19:57] I just got it to work (made some changes in the code) [02:20:05] kk [02:20:22] I have another question [02:21:53] simonrvn: Quote from Mediawiki website "ImageMagick renders SVGs, but very buggily. Also slow. This is the default, but avoid it if possible." [02:22:23] A main option which I haven't figured out yet - how would I make sure that the wiki is private (as in no one would be able to acssess it except me) and how would I be able to get certain people to accsess it and grant them the neccasery permissions to help me edit stuff? [02:23:05] $wgGroupPermissions['*']['create'] = false; [02:23:05] $wgGroupPermissions['*']['createaccount'] = true; [02:23:05] $wgGroupPermissions['*']['read'] = false; [02:23:05] $wgGroupPermissions['*']['edit'] = false; [02:23:14] Add those lines, that should do the trick [02:23:29] oh [02:23:30] hold on [02:23:39] Line 2 change to false [02:23:50] Kodos2: ping [02:24:04] ? [02:24:16] ping? [02:24:29] Just wanted to make sure you read the above [02:24:46] ping is an attention getting word... [02:25:02] oh ok [02:25:51] so after adding that code no one would be able to enter the wiki ? how would I be able to add people ? what address would I give them and how would I define their user's permissions? [02:27:13] what version of php does mediawiki runs now? [02:27:30] 03(mod) wgSpamRegex addition request: anontalk.com - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16597 +comment (10mikelifeguard) [02:27:38] xah_lee, what you mean? [02:28:05] nn123645 what version of php do i need to install? [02:28:07] wampserver 2.0 [02:28:20] as long as its php 5 it should work [02:28:39] nn123645: php 4.x not ok? [02:28:43] thx. [02:29:01] how would I be able to add people ? what address would I give them and how would I define their user's permissions? [02:29:18] PHP 4 support is ending in a few months, I wouldn't recommend using it [02:29:18] JKodos2: Yes [02:29:27] kk [02:29:38] You can add people by going to Special:Createaccount [02:30:07] Address is up to you or them [02:30:24] You would do permissions the defualt way [02:31:31] how do I know what address to give them? [02:32:14] ask them [02:33:03] is it something like http:///main menu.htm ? [02:34:44] Kodos2: Im confussed [02:35:26] oh [02:35:28] the address [02:35:31] yea [02:35:33] just clicked [02:36:02] ? [02:36:46] yeah [02:36:53] itoh [02:37:04] it will probably be just ur ip [02:37:19] depends if you put mediawiki in asub folder [02:37:36] make sure you setup port forwarding on ur router [02:39:21] only one problem ... I'm not quite sure what the last two lines you write mean exactly [02:41:33] sigh [02:41:37] giv me a sec [02:41:47] ok [02:44:51] Ok [02:44:57] now that i have got coffee [02:45:03] and have braced my self ;) [02:45:15] Do you know what the brand of your router is [02:47:38] microsoft [02:48:06] mn-500 [02:49:39] k [02:49:41] giv me a sec [02:49:47] (getting manual) [02:51:54] Ok, first of all you need to find out the IP of your machine [02:52:17] So go Start >> Run [02:52:17] ok [02:52:23] and type in cmd [02:52:33] Command prompt should open ip [02:52:33] ok [02:52:35] up^ [02:52:42] then type in ipconfig [02:52:57] and press enter [02:53:38] You should see a line that says "IP Address ............ " [02:53:45] whats the IP? [02:53:54] is it the ipv4 address or the subnet mask or the deafult gateway? [02:54:03] ipv4 address [02:54:07] ok [02:54:22] what should I do with it? [02:54:24] So yeah, write that down somewhere [02:54:32] ok [02:55:05] now open up your browser and go to http://192.168.2.1 [02:55:43] It will ask you for a password that you would have set when you set it up [02:55:55] if you dont know the password, try admin [02:56:53] when your in let me know [02:57:44] done [02:58:17] ok click on "Securty" [02:58:27] then click "Port Forwarding" [02:58:30] ok [02:58:45] then click "Set up persistent Port Forwarding" [02:59:26] In the Description box type something like "HTTP" [02:59:47] In the inbound port box type 80 [03:00:06] ok [03:00:11] in the type box select TCP [03:00:27] in the Pricate IP Address box, type the IP that your wrote down earlier [03:00:43] in the Private port box, type 80 [03:00:49] Then click apply [03:01:11] ok [03:01:18] all done? [03:01:39] yes [03:02:20] Have you got the website server running atm? [03:02:20] 03tstarling * r44556 10/trunk/extensions/TrustedXFF/ (trusted-hosts.txt update): [03:02:21] * Added an update script [03:02:21] * Updated TPG proxies, removed some dead BC proxies, added TeliaSonera, split out IWF proxies by ISP [03:03:36] Kodos2: Have you got the website server running at the momment? [03:03:39] I got is wampserver 2 [03:03:46] yes [03:04:01] Its it running on the machine that you are currently on? [03:04:06] yes [03:04:23] can you access it by going to localhost? [03:05:15] yes [03:05:51] Can you pm me your external IP so that I can check that it is working? [03:07:37] 03(mod) Allow anon users to define which version should be displayed by default - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14321 (10JSchulz_4587) [03:10:11] ok [03:10:33] So i cant connect atm and im guessing its the windows firewall [03:11:21] Kodos2: Can you join #windows [03:11:40] ok [03:11:54] AaronSchulz: can't you just say to ^ bug, "anon preferences will break cache integrity, no way in hell" [03:12:05] how would i set $wgLogo? [03:13:16] lol, that too [03:13:26] as I understand it, in order of importance (from highest to lowest) to Wikimedia techs: millions in donations for tea/coffee/coke > cache integrity > site usability > site content > site requests > users being happy [03:13:43] Splarka: you're welcome to comment [03:14:05] bah [03:14:22] I don't know enough about it [03:14:35] tell me.. can any anon viewer view the most current version by clicking history and then the top date? [03:14:52] that is, they are always two clicks away from it? [03:16:57] yeah [03:17:42] are they ever one click away from it? [03:17:55] like, in the toolbox.. 'latest draft' or such [03:18:07] 03(mod) Allow anon users to define which version should be displayed by default - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14321 (10JSchulz_4587) [03:21:02] I am wondering if you could have some sort of cascading parameter (although I am not aware of such in any core features)... a sort of &showunstable=true parameter link they can click, which then turns all local wikilinks into having the same [03:21:45] cache would be maintained (if messily), you could have obtrusive warnings about the viewer seeing pages as drafts... [03:24:20] eg, if anon user was on /wiki/Foo, and clicked unstable mode, they'd end up on /w/index.php?title=Foo&showunstable=true, and all local links to ns >= 0 (or maybe just content namespaces) would then be changed to /w/index.php?title=TITLE&showunstable=true (dirrrrrty) [03:43:22] 03vyznev * r44557 10/trunk/phase3/includes/filerepo/RepoGroup.php: Followup to r44525: remove the broken $flags parameter from RepoGroup::findFiles() too. [03:49:46] I was wondering how to have the Main Page point to a different location, the FAQ wasn't to specific [03:50:43] you mean the logo and navigation link? edit [[MediaWiki:Mainpage]] [03:51:05] (also: default page if no title specified, default redirect target after some actions, etc) [03:52:21] Mk [03:53:20] 03aaron * r44558 10/trunk/phase3/includes/filerepo/ (File.php LocalFile.php): Make boundary inclusion optional in getHistory() [03:57:08] hi! [03:59:09] Hi [04:02:49] 04(REOPENED) Time based anti-vandalism feature - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=4397 +comment (10ian.woollard) [04:10:21] quick question here: I imported pages (xml data) and they don't seem to show up as new articles in the little article counter. Any one has an idea why that is? [04:10:23] 03(FIXED) Time based anti-vandalism feature - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=4397 +comment (10mikelifeguard) [04:14:15] Eight minutes. Nice. [04:14:34] Mathieu1: You need to run a maintenance script I think. [04:14:39] To update the database fields. [04:14:49] How many pages did you import? [04:15:57] MZMcBride: about 300 [04:16:13] I think rebuildAll.php would probably fix all of that. [04:16:16] I think. [04:16:30] Assuming you're able to run maintenance scripts directly. [04:16:37] yep [04:17:07] rebuildall.php * [04:17:55] Or updateArticleCount.php [04:17:57] MZMcBride: do you know what the script is? [04:17:59] One of them. [04:18:05] thank you! [04:18:12] With only 300 pages, any of them will be quick. [04:18:17] Run them all! ;-) [04:19:02] do you know the path to these php pages by any chance? I don't have access to the files on the server right now :( [04:19:24] It's just... [04:19:35] php /maintanence/rebuildall.php [04:19:37] Or whatever. [04:19:42] From the wiki directory. [04:19:46] I see. [04:19:54] First slash may be unnecessary. [04:21:36] 03(mod) Do not hide #mw-commentbox - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16480 (10Wiki.Melancholie) [04:23:20] MZMcBride: my w/maintenance/updateArticleCount.php doesn't work :( Do you think I typed it wrong? [04:23:45] Which directory did you start from? [04:23:55] And did you set the proper information in AdminSettings.php ? [04:24:32] the whole path is: en.mydomain.com/w/maintenance/updateArticleCount.php [04:24:49] I didn't know I had to set something in AdminSettings.php :( [04:24:59] You have to run maintenance scripts from the command line. [04:25:09] They cannot be run from the Web. [04:25:16] oh! the SSH? [04:25:22] Yeah. [04:25:24] oki [04:25:30] cd to the w directory. [04:25:38] 03(mod) Wikimedia dumps do not contain any information about the ( flagged) article status - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16546 (10Wiki.Melancholie) [04:25:42] Setup your AdminSettings.php file. [04:25:47] And then run the maintenance scripts. [04:28:47] ffs... why the heck does PHP need 75 different array functions? [04:28:54] Perl has, what, 11? [04:29:11] hey more is better :) [04:29:16] MZMcBride: I'm not very familiar with running script using putty... Would you know the way after I just signed in as the root? [04:29:36] You need to edit the file called AdminSettings.sample. [04:30:01] Change its name to AdminSettings.php and put the correct information inside of the file (there are notes in the file). [04:30:17] Then simply run "php maintenance/scriptname.php" [04:31:35] hmm.. I don't have access to the AdminSettings.php file tonight. I guess I'll need to wait tomorrow. [04:32:08] what was the cd to the w directory? would you know the command line by any luck? [04:32:44] cd == change directory [04:32:50] It's how you move from one folder to another. [04:32:54] cd w [04:33:09] that might be usefull :) [04:34:04] it says: bash: cd: w: No such file or directory [04:34:18] Type "ls" [04:34:36] bash: Is: command not found [04:34:36] That tells you all the directories and files inside your current directory. [04:34:47] L [04:34:50] ls [04:35:15] "pwd" will tell you your present working directory (where you are currently). [04:35:19] there we go! [04:35:35] en index.html invitation [04:35:36] images index.html.LCK public_html [04:36:18] So perhaps your directory is named "en", not "w"? [04:36:31] en is my subdomain [04:36:39] en.domain.com/w [04:36:43] Ahh. [04:36:51] cd en/w [04:37:29] should I do "cd == en/w" [04:37:43] because bash: cd: en/w: No such file or directory [04:38:15] Try "cd en" [04:38:22] Then try "ls" [04:38:36] there we go! [04:39:33] I see AdminSettings.sample in there [04:41:33] MZMcBride: should I run the maintenance script php /maintanence/rebuildall.php? [04:42:02] Helps if you spell maintenance correctly. [04:42:08] And I believe the first / is unneeded. [04:42:27] And you probably should try updateArticleCount.php first [04:44:06] Permission denied ?!? [04:45:05] You told me to first make some changes in the adminsetting.php right? [04:47:00] The first / isn't unneeded, it's outright wrong. It will refer to a directory maintenance/ in the root directory (which probably doesn't exist). [04:47:55] hi Simetrical! [04:48:24] here's what I get: bash: maintenance/updateArticleCount.php: Permission denied [04:56:06] 03(mod) Global user style - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13953 +comment (10Wiki.Melancholie) [04:57:44] 03(mod) Global transclusion/userpage - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16575 +comment (10Wiki.Melancholie) [04:58:36] Simetrical: could having a subdomain mess up the script? [04:58:46] i tried several times and came back in /root [04:59:34] 03pinky * r44559 10/trunk/extensions/Configure/Configure.settings-ext.php: Change InerwikiList --> InterwikiList in Configure.settings-ext.php [05:04:17] :/ [05:04:50] :P [05:07:11] Hey ^demon. [05:07:28] <^demon> Evening. [05:18:05] 03(mod) Code Review buttons can be considered "confusing" - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16639 (10p858snake) [05:18:08] 03(mod) Code Review buttons can be considered "confusing" - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16639 (10innocentkiller) [05:22:32] *^demon kicks wikibugs [05:28:29] good night all [05:30:13] !Interwiki Kodos2 [05:30:13] --mwbot-- Interwiki links are links between wikis. For instance, on Wikibooks, [[w:Article name]] or [[Wikipedia:Article name]] will link to a Wikipedia article. To set up interwiki links, see and . For easy configuration, use [05:33:28] 03(mod) SVGZ (gzipped SVG) support - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=4947 +comment (10steveb05+wikibugs) [05:35:07] Hello, I just posted that SVGZ patch to bugzilla. Would anyone like to help test it? [05:39:52] why compress them in mediawiki? just have the web server do it, like every other text file [05:41:14] Many reasons. Web servers aren't the only place where e.g. Wikipedia is stored - also dbdumps, Cdroms, et al. [05:41:46] I think it makes sense to leave the files compressed until the final rasterization [05:47:08] yksinaisyyteni: I gots a question, uploaded files are, before being checked for stuffs (and interstitial warnings/errors/crap), sent to temp limbo, right? images/temp/x/xy/dateTimestamp!imagename or something [05:47:22] something like that [05:47:29] how long are they left there? does MediaWiki have any say in the matter or does the system clean them whenever it wants? [05:47:48] assuming that the user doesn't click one of the two buttons, that is [05:47:59] [Ignore warning and save file anyway] [Cancel upload and return to the upload form] [05:49:08] <^demon> grumpy_: Might I suggest testing for function_exists( 'gzopen' ) with $use_gzip? Needs to fail nicely if PHP wasn't compiled with zlib. [05:49:25] good call [05:50:06] (although I can't imagine why someone would compile their PHP without it, it is conceivable) [05:51:03] *Vyznev pokes CIA-8 again [05:51:25] <^demon> However, what happens if svgz is passed but PHP lacks gzip support? Is it going to try and parse and barf? [05:51:28] for some reason, the bot doesn't seem to want to report my commits :| [05:51:40] <^demon> Vyznev: Wikibugs is lagging too :( [05:52:53] demon: if a zlib check fails, mediawiki should fail gracefully to a "not well formed" user error [05:54:00] <^demon> True, yes. [05:54:36] *Splarka wonders if river suddenly found 8 years worth of aborted upload temp files on wm [05:54:43] heh [05:54:46] Splarka: you'd have to ask brion [05:54:56] i didn't notice a large number of files in temp/ when i was copying it [05:55:37] odd, so I wonder how it knows when to delete them [05:55:54] and... if naughty things can be done to them in the meanwhile... [05:58:24] chuck ping [05:59:42] 03(NEW) UTF-8 in commit message confuses CodeReview - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=16640 minor; normal; MediaWiki extensions: CodeReview; (nospam) [06:00:46] 03(mod) SVGZ (gzipped SVG) support - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=4947 +comment (10steveb05+wikibugs) [06:00:47] <^demon> Vyznev: I think that's SVN not liking it? My e-mail shows it broken too. [06:01:02] <^demon> Are svn commit messages utf-8? [06:01:10] well, ViewVC shows it correctly at least [06:01:49] <^demon> :\ [06:03:09] 03(mod) UTF-8 in commit message confuses CodeReview - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16640 (10nospam) [06:17:36] 41 previously passing test(s) now FAILING! :( [06:21:37] How does one commit patches once they're approved? [06:22:29] rar JSharp and Splarka [06:23:02] gumby_: usually a developer does that [06:23:32] rar [06:23:49] So should I just sit tight until a dev comments on my bugreport? [06:24:16] that can take years if it is a feature/enhancement [06:24:55] My point exactly [06:26:48] channel steve ballmer [06:26:55] http://www.mediawiki.org/wiki/Developers_Developers_Developers_Developers!_Developers_Developers_Developers_Developers!_Developers_Developers_Developers_Developers!_Developers_Developers_Developers_Developers!?curid=16019 [06:28:23] I see wat u did thar... [06:29:19] I was gonna do http://www.mediawiki.org/wiki/Developers?_Developers_Developers_Developers!... too [06:29:32] *gumby_ throws chair [06:30:25] gumby_: hit brion (spelling??) with a baseball bit till he does it the next time he's in here [06:32:22] thx for the help, I'll just try to catch one of the devs I guess [06:32:34] *^demon looked it, but he's not Brion [06:32:45] <^demon> And knows nothing about SVGZ :p [06:33:53] *Vyznev looked at it too [06:34:25] seems pretty reasonable, at a glance, but I'm not very familiar with dile type detection either [06:35:01] Thanks for the help guys. Just my little pet project, that's all. [06:35:06] hmm... you seem to have a stray xml_parser_free( $parser ) there? [06:35:28] *Dantman wonders if he should do some sort of developer task trade [06:35:59] ie: create one feature in PHP if someone helps him debug one of his JS libraries into completion [06:36:47] Vyznev: methinks it should only free() once, since if it's misformed it'll return immediately [06:37:26] ah, right, I missed the return there [06:38:11] now that I look at it, it's not very succinct. Could just set $this->wellformed = false inside that block [06:41:50] Thanks all, I'll come back later. Bedtime here. [06:45:44] *^demon throws rocks at ForeignApiRepo [06:45:53] :O [06:46:10] 04(REOPENED) Time based anti-vandalism feature - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=4397 +comment (10ian.woollard) [06:55:15] 03(mod) problem on Persian title - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=16638 (10amirali.b) [06:55:34] 03demon * r44561 10/trunk/phase3/includes/filerepo/ForeignAPIRepo.php: FindBySha1 now works in ForeignApiRepo :) [06:56:46] 03(mod) SVGZ (gzipped SVG) support - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=4947 +comment (10nospam) [07:02:51] 03aaron * r44562 10/trunk/phase3/includes/filerepo/LocalFile.php: Expand ASC condition as it should be [07:02:59] 03demon * r44563 10/trunk/phase3/includes/filerepo/ForeignAPIRepo.php: Add some sanity checking. Return nicely if Http::get fails (in general) on fetchImageQuery(), and also if our query returns poorly in findBySha1(), similar to queryImage(). [07:10:55] 03aaron * r44564 10/trunk/phase3/includes/ImagePage.php: (bug 12645) File histories should be paged [07:11:02] 03(FIXED) File histories should be paged - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=12645 summary; +comment (10JSchulz_4587) [07:12:16] 03aaron * r44565 10/trunk/phase3/includes/ImagePage.php: Remove comment [07:23:49] uh oh, someone seems to have broken a bunch of image-related parser tests [07:23:55] *Vyznev hopes it wasn't him [07:24:21] Vyznev: I think it was [07:24:40] see comments on r44560 [07:26:11] 03tstarling * r44566 10/branches/REL1_13/phase3/ (4 files in 3 dirs): Backported IE simulation from trunk as of r44505. [07:31:55] okay, that's weird [07:31:59] php maintenance/parserTests.php --regex='Simple image' passes [07:32:12] php maintenance/parserTests.php --regex='(thumb|Simple) image' fails [07:33:05] somehow, running the "Template with thumb image" test first causes the other two tests to fail [07:34:23] I added the file cache to the parser after some pages on enwiki that used the same images hundreds of times stopped loading [07:38:29] 03vyznev * r44567 10/trunk/phase3/maintenance/parserTests.inc: Fix regression in parserTests.inc caused by r44560. [07:42:27] 41 previously failing test(s) now PASSING! :) [07:43:18] if you can't fix the parser, fix the tests... [07:44:17] in this case, though, it was the tests that I broke and not the parser [07:44:45] Interesting question, how does one get at the temporary file saved when an upload is interrupted for an interstitial (a warning message, etc) for display in an interstitial warning? [07:50:18] 03tstarling * r44568 10/branches/REL1_13/phase3/ (RELEASE-NOTES includes/DefaultSettings.php): [07:50:18] * Less euphemistic release notes. Aiming for December 15 release. [07:50:18] * Added text/scriptlet to MIME type blacklist as in trunk. [08:04:13] is there any way to make it so linebreaks are preserved? i always have to add an extra
| example: http://rafb.net/p/JcA84U44.html | MW version: 1.12.1 (i think... it seems the version page is broken, it says "1.14alpha") [08:06:34] wikicode is much like html, dynamic wrapping is natural for text flow [08:07:04] two blank lines however starts a new paragraph

(or other block level element) [08:07:37] hm, ok [08:07:38] And three makes you look stupid [08:07:42] depending on what you want the linebreaks for, there are several different solutions. what do you want the linebreaks for? [08:08:01] showing source code? 

 (core) or  (geshi extension) are good
[08:08:29] 	for indented lists? : or 
, * or 
  • , # or 
    1. , etc
[08:08:45] 	i generally use a single linebreak for lines of related-but-different text
[08:09:09] 	most people'd probably just start new paragraphs then, unless you don't want the margin
[08:10:14] 	eh, oh well, it's not too big of a deal; i've just always been a bit annoyed at that >_>
[08:10:21] 	there is the  extension, which basically preserves newlines naturally, while allowing proportional fonts, wikicode parsing, and wrapping
[08:12:38] 	there's also a problem with the version page (or something related to it), because it displays "1.14alpha" for the MW version, when I know we're using 1.12.1
[08:14:01] 	it seems more likely that you're using a copy from svn
[08:26:32] 	well i know i'm using 1.12.1, that's what i installed and i haven't upgraded it yet
[08:26:36] 	so...
[08:26:58] 	then the only explanation is that cosmic rays cause the text "1.12.1" in your installation to morph into "1.14alpha"
[08:28:49] 	>_>
[08:31:17] 	03tstarling * r44569 10/branches/REL1_13/phase3/includes/api/ (ApiFormatBase.php ApiMain.php): Backported r43470, redundant fix for Rem1.
[08:38:20] 	where's the version number stored, anyway?
[08:38:29] 	includes/DefaultSettings.php
[08:38:46] 	03tstarling * r44570 10/branches/REL1_13/phase3/includes/filerepo/ (. FSRepo.php): Backported r43625 again, apparently it got left out of r43669 somehow.
[08:39:12] 	thanks
[08:42:04] 	will you edit that to say 1.12.1, so you can pretend that's what you're running? ;)
[08:43:18] 	03(NEW) Errors in centralnotice.js - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16641 major; normal; Wikimedia: General/Unknown; (apper)
[09:25:10] 	Hey, I'm an admin at http://es.wikifur.com, I found spelling error in the Spanish Bad title text, ¿who may help me?
[09:28:38] 	RichardWolfVI: uh... clcik adit and fix it?
[09:29:23] 	err, click "edit".
[09:29:34] 	Duesentrieb: It's been already done in the wiki, of course, but I'd like that fixed in the
[09:29:36] 	man, what is it with my typing lately?
[09:29:55] 	*I'd like it fixed in MediaWiki
[09:30:24] 	hm... the right place for that is betawiki
[09:30:35] 	http://translatewiki.net/wiki/Main_Page
[09:30:40] 	they manage all the translations
[09:30:55] 	ok, thanks for the info
[09:45:55] 	03siebrand * r44571 10/trunk/phase3/languages/messages/ (76 files): Localisation updates for core messages from Betawiki (2008-14-10 10:36 CET)
[09:50:21] 	03siebrand * r44572 10/trunk/extensions/ (105 files in 90 dirs): Localisation updates for extension messages from Betawiki (2008-14-10 10:36 CET)
[09:53:17] 	03(NEW) Lift restriction on unregistered page creation in project space - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16642 15enhancement; normal; Wikimedia: General/Unknown; (m.s.jones)
[09:57:20] 	03(mod) Lift restriction on unregistered page creation in project space - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16642  (10m.s.jones)
[09:59:59] 	morning
[10:00:04] 	03siebrand * r44573 10/branches/REL1_13/phase3/maintenance/language/messages.inc: Update messages.inc
[10:12:06] 	03vyznev * r44575 10/trunk/phase3/maintenance/parserTests.txt: better parser test coverage of {{ns:}}
[10:12:08] 	huomenta
[10:12:09] 	03river * r44574 10/trunk/tools/ts-specs/TSphp.spec: 5.2.8
[10:35:46] 	03siebrand * r44576 10/branches/REL1_13/phase3/languages/messages/ (118 files): 
[10:35:46] 	Localisation updates from Beatwiki
[10:35:46] 	* Backport compatible localisations to REL1_13 (2008 new translations, existing messages updated)
[10:36:30] 	beat it to death wiki :o
[10:36:42] 	oi
[10:49:19] 	14(DUP) Display a table of contents on Special:Watchlist/edit - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=2477  +comment (10raimond.spekking)
[10:49:20] 	03(mod) Add a TOC to Special:Watchlist/edit - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16635  +comment (10raimond.spekking)
[10:51:57] 	hey, how do i do some kind of box where everything inside is parsed literally? (much like the [code] in many bbcode dialects)
[10:52:26] 	kqr: 
      
[10:52:51] 	wow you were quick
[10:52:54] 	but thanks anyway
[10:52:57] 	just what i wanted
[10:52:57] 	(:
[10:53:07] 	or, if you want to get fancy, install the syntax highlight extension and use 
[10:53:20] 	x]
[10:53:24] 	geshi or whatever its called ^.^
[10:53:26] 	not needed :P
[10:56:01] 	thats what i used to say but when i'm coding now i <3 it
[11:07:14] 	03danny_b * r44577 10/trunk/extensions/CleanChanges/CleanChanges_body.php: * + missing required alt attribute; currently empty, but it should contain something though because of accessibility
[11:13:55] 	03siebrand * r44578 10/branches/REL1_12/phase3/languages/messages/ (111 files): 
[11:13:55] 	Localisation updates from Betawiki
[11:13:55] 	* Backport compatible localisations to REL1_12 (1455 new translations, existing messages updated)
[11:15:47] 	03nikerabbit * r44579 10/trunk/extensions/CleanChanges/ (CleanChanges.i18n.php CleanChanges_body.php): * Show a useful messages instead of nothing
[11:24:52] 	03(NEW) Some diffs are escaped thus appear in single line - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16643 major; normal; MediaWiki extensions: CodeReview; (danny.b)
[12:33:47] 	does someone know of a good tool for multiwiki user management?
[12:34:38] 	basically I'd need some way to see who has access/edit rights to 4 different mediawikis, they share most users/permissions
[12:53:21] 	share the user databases
[13:31:50] 	Is it possible to share the users and their groups for different wikis?
[13:36:54] 	Wuzur: using shared databases
[13:38:13] 	[[Manual:$wgSharedDB]]
[13:42:30] 	so adding $wgSharedTables=array( 'user' , 'user_groups'); $wgSharedDB='dbname'; should work?
[13:43:37] 	i've never done it before so i have no idea
[13:43:52] 	hmm, ok, then I start testing that
[13:45:50] 	anyone known an extension which will let me overlay text and vector images on a graphic?
[13:45:52] 	erm
[13:46:00] 	vector graphics on an image?
[13:46:19] 	I figure it would probably have to use imagemagic on the back end
[13:46:24] 	just wondering thanks
[14:25:50] 	nm
[14:25:59] 	I'll make one I think
[14:26:51] 	03(WONTFIX) Time based anti-vandalism feature - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=4397  +comment (10bretthillebrand)
[14:27:11] 	since I want to overlay multiple SVGs onto a single PNG backgroup and render it as a PNG
[14:27:21] 	.
[14:27:22] 	.
[14:27:44] 	Is there a special page which will show me all the red links in my MW?
[14:28:33] 	try Special:WantedPages
[14:35:04] 	03(mod) SVGZ (gzipped SVG) support - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=4947  +comment (10Simetrical+wikibugs)
[14:43:10] 	siebrand, happy birthday, btw
[14:43:18] 	04(REOPENED) Time based anti-vandalism feature - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=4397  +comment (10ian.woollard)
[14:43:36] 	kibble: thanks :)
[14:45:52] 	03(WONTFIX) Time based anti-vandalism feature - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=4397  +comment (10maxsem.wiki)
[15:14:54] 	03rotem * r44580 10/trunk/phase3/languages/messages/MessagesHe.php: Localization fix.
[15:15:35] 	03rotem * r44581 10/branches/REL1_13/phase3/languages/messages/MessagesHe.php: Localization fix.
[15:15:54] 	03rotem * r44582 10/branches/REL1_12/phase3/languages/messages/MessagesHe.php: Localization fix.
[15:23:29] 	Hello, I'm having a problem with wgAPIPropModules. Is it possible to add custom properties?
[15:38:32] 	03(mod) Lift restriction on unregistered page creation in project space - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16642  +comment (10matthew.britton)
[15:39:18] 	Help. I install two different versions of MediaWiki (1.11.0 and 1.12.0) onto my computer at the same time. Now the version I don't want to use (1.12.0) has taken over. Whenever I visit my main page, the links point to URLs for the removed Mediawiki software.
[15:39:45] 	Something in my database has changed, but I don't know what that is.
[15:39:49] 	mumia-w-19: Change $wgArticlePath in LocalSettings.php
[15:40:01] 	RoanKattouw: thanks, I'll try that.
[15:41:59] 	RoanKattouw: There is no such setting in my LocalSettings.php
[15:42:10] 	Oh wait
[15:42:15] 	Well it's $wgArticleSomething
[15:42:19] 	I think something changed in my database.
[15:42:25] 	I doubt it
[15:42:39] 	03(mod) problem on Persian title - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=16638  +comment (10matthew.britton)
[15:43:00] 	No, there is nothing like "wgArticle" in my LocalSettings.php
[15:43:35] 	Hmm. There should be
[15:43:41] 	mumia-w-19, paste it to a pastebin (minus DB settings).
[15:43:57] 	See if $wgScriptPath and $wgScript look sane
[15:44:58] 	03(mod) API protection query returning wrong result - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16629  +comment (10roan.kattouw)
[15:50:54] 	03catrope * r44583 10/trunk/phase3/includes/api/ApiQueryRecentChanges.php: API: Fix broken block expiry calculation in list=recentchanges
[15:50:55] 	03(FIXED) Block expiry time for list=recentchanges - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16588  +comment (10roan.kattouw)
[15:52:20] 	Here is my LocalSettings.php: http://pastebin.mozilla.org/591942
[15:53:22] 	mumia-w-19: If your links point to the wrong wiki, you probably have to change $wgScriptPath
[15:53:22] 	There is no $wgScript, but $wgScriptPath looks good; it's empty.
[15:54:19] 	Can you give us a link to your wiki?
[15:54:44] 	I'll try; just a minute.
[15:55:47] 	To mess up my wiki; I didn't change anything in LocalSettings.php. I simply visited this URL: http://localhost:81/mediawiki/config/index.php
[15:59:57] 	03(mod) Admins don' t see when page creations match against the blacklist - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13780  +comment (10matthew.britton)
[15:59:58] 	Evidently something in config/index.php changed my database so that wiki links would like to the new mediawiki installation.
[16:00:13] 	I'm still trying to make my wiki public.
[16:00:22] 	mumia-w-19: That information is not in the database, as I said before
[16:00:27] *mumia-w-19 	continue looking at the apache configuration.
[16:00:29] 	okay
[16:03:00] 	domas, Simetrical: Could you guys do me a favor and look into the MySQL craziness I found a week ago? http://rafb.net/p/EQaaYe31.html
[16:03:29] 	RoanKattouw, didn't I already explain why it makes sense?
[16:03:39] 	DISTINCT will often add a filesort.
[16:03:46] 	Yeah I got that
[16:03:57] 	So what's the problem?
[16:04:03] 	Except now I'm only SELECTing *one* field (page_title) that I'm also ordering by
[16:04:14] 	Basically I've got SELECT DISTINCT page_title FROM .... ORDER BY page_title
[16:04:32] 	Ah.
[16:04:43] 	Where are you testing this?
[16:04:51] 	Now I've proven my database ignorance many times over :P but *that* shouldn't need a filesort, right?
[16:04:55] 	Oh, I see.
[16:04:56] 	In theory, no.
[16:04:59] 	But MySQL is stupid.
[16:05:28] 	Simetrical: Just running some EXPLAINS on 5.0.51a-community-nt, really small DB
[16:05:37] 	So I don't know if it'll actually cause a slow query in practice
[16:05:57] 	Oh, it will, I'm pretty sure of that.
[16:06:04] 	...
[16:06:19] 	Hmm, MySQL is even more stupid than I thought
[16:06:28] 	It is pretty stupid, yeah.
[16:07:15] 	There should be some way to do it.
[16:07:17] 	Simetrical: Does the toolserver run 5.0 or 5.1?
[16:07:21] 	5.0.
[16:07:30] 	Right. Me too
[16:07:32] 	5.1 wasn't GA until just now.
[16:07:52] 	And the servers still run 4.1, right?
[16:07:57] 	You just want main-namespace pages that have langlinks?
[16:07:58] 	4.0.
[16:08:00] 	Yes
[16:08:23] 	Without the DISTINCT it's nice and efficient but returns the same page like a hundred times
[16:08:42] 	This seems to work: EXPLAIN SELECT STRAIGHT_JOIN page_title FROM page, langlinks WHERE page_id = ll_from AND page_namespace =0 ORDER BY page_title
[16:08:45] 	At least on the toolserver.
[16:08:58] 	Er, wait.
[16:09:01] 	That doesn't have DISTINCT.
[16:09:14] 	Adding that re-adds the problem.
[16:09:17] 	Yeah WFM too, but needs DISTINCT as you said
[16:09:23] 	Yup
[16:10:09] 	Dunno, ask domas or #mysql.
[16:14:08] 	03(FIXED) Default parameter is broken - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16184  +comment (10ran.arigur)
[16:22:46] 	Simetrical: This query should be functionally equivalent, right?   SELECT STRAIGHT_JOIN page_id, page_namespace, page_title FROM page, langlinks WHERE page_id = ll_from GROUP BY page_namespace, page_title ORDER BY page_namespace, page_title
[16:23:03] 	Ah, GROUP BY.
[16:23:07] 	That's a good idea.
[16:23:16] 	That should get you uniqueness, yeah.
[16:23:22] 	Allows me to DISTINCT on ns,title while also selecting pageID
[16:23:23] 	And probably kill the filesort.
[16:23:34] 	It does kill the filesort. Does it get me uniqueness?
[16:24:03] 	It should, yes.
[16:24:59] 	Hello, my wiki is finally viewable from the outside: http://uxncc1701c.dyndns.org:81/index.php/Main_Page
[16:25:09] 	fails on postgresql, though; page_id would have to be included in the GROUP BY or in an aggregate function
[16:25:24] 	anomie: WTH? Why?
[16:25:46] 	RoanKattouw- MySQL allows the deviation from the SQL standard, IIRC
[16:25:58] 	The last link, User:Wade-user, links to the wrong place.
[16:27:01] 	http://uxncc1701c.dyndns.org:81/mediawiki/index.php/User:Wade-user should instead be this: http://uxncc1701c.dyndns.org:81/index.php/User:Wade-user 
[16:27:08] 	Well OK, adding page_id to the GROUP BY doesn't seem to hurt
[16:27:14] 	IOW, the /mediawiki/ should not be there.
[16:27:22] 	RoanKattouw, you could write MAX(page_id) or something, if you like.
[16:27:47] 	Well I guess it's not necessary, since the variant with page_id in the GROUP BY doesn't filesort either
[16:27:48] 	That would make it distinct (although really it's well-defined anyway because of the uniqueness constraints, SQL doesn't like that).
[16:28:24] 	RoanKattouw, that's the kind of thing that might fail in MySQL 4.  It's probably figuring out that grouping by the page_id is redundant.
[16:28:41] 	Hmm
[16:28:51] 	So you're saying I should select MAX(page_id) AS garbage ?
[16:29:19] 	No, SELECT STRAIGHT_JOIN MAX(page_id) AS page_id, page_namespace, page_title FROM ...
[16:29:40] 	Simetrical: Right.... but that'd still give me the right page_id, right?
[16:29:49] 	Yeah.
[16:29:49] 	Not the highest page id for every row?
[16:29:58] 	MySQL just picks an arbitrary value (probably first or last row, I don't recall) when the GROUP BY doesn't specify the column
[16:30:03] 	It will give you the highest page_id for every row.
[16:30:07] 	Which happens to be the only one.
[16:30:12] 	Could do MIN(page_id) too if you like.
[16:30:22] 	Right
[16:30:23] 	They're all the same here, so it really makes no difference.
[16:30:43] *anomie 	once wrote aggregates "first" and "last" for use in PG
[16:30:46] 	But that's gonna be programatically difficult, since the array of stuff to select may be variable
[16:31:33] 	Which means I've got SELECT page_namespace, page_title, page_id, /* possibly more page fields here */ FROM page ...
[16:32:00] 	03(NEW) Allow users to be blocked from editing non-talk pages only - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16644 15enhancement; normal; MediaWiki: Blocking; (matthew.britton)
[16:32:58] 	RoanKattouw, so either 1) decide pgsql can go screw itself, or 2) put MAX() on everything.  (2) will probably hurt performance in MySQL.
[16:33:04] 	Actually it might even with one page_id.
[16:33:07] 	Simetrical: Wait a minute. Would MySQL 4 even have a problem there? There's a unique index on (ns, title), so it *knows* (or at least it should) that grouping by ns, title yields unique values for all other page fields
[16:33:20] 	Correct.  MySQL doesn't care.
[16:33:25] 	Wait, what?
[16:33:29] 	Oh, you mean what I said before.
[16:33:34] 	Depends, MySQL is stupid.
[16:33:35] 	Well not with the pgsql thing
[16:33:40] 	Yeah, I guess
[16:33:59] 	Anyone have a MySQL 4 install lying around?
[16:34:07] 	I doubt MySQL "knows" the other fields are unique, I believe it just picks some row at random and uses the values from that.
[16:34:32] 	anomie, no, he's talking about something else.  GROUP BY page_namespace, page_title, page_id not causing a filesort.
[16:34:35] 	anomie: It should know because of the unique index
[16:34:49] 	Simetrical- Oh, sorry
[16:34:58] 	Should only be possible by inference from the unique index on (ns, title).
[16:35:04] 	After doing GROUP BY page_namespace, page_id , all page_* fields are identical because of the unique index on (ns,title)
[16:35:22] 	So grouping by page_whatever won't change anything
[16:35:27] 	Logically, no.
[16:35:28] 	RoanKattouw- It could know, but I doubt it does because it allows the same thing when the results aren't unique.
[16:35:30] 	Implementationally, who knows?
[16:35:49] 	Domas may know
[16:37:24] 	domas?
[16:42:28] 	Can anyone help me? I have my wiki up: http://uxncc1701c.dyndns.org:81/index.php/Main_Page
[16:46:56] 	Hello?
[16:47:56] 	03(mod) problem on Persian title - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=16638  (10amirali.b)
[16:50:34] 	mumia-w-19: Purging the page fixed it
[16:50:46] 	mumia-w-19: Are you by any chance running two wikis on the same DB?
[16:50:59] 	Unfortunately, yes :-(
[16:51:23] 	How do I purge a page?
[16:51:34] 	mumia-w-19: http://uxncc1701c.dyndns.org:81/index.php?title=Main_Page&action=purge
[16:51:45] 	Thanks :-)
[16:51:50] 	Same URL as you get for edit, except with a different action=
[16:52:05] 	Can I purge all pages?
[16:52:20] 	If you have database access, yes
[16:52:52] 	TRUNCATE objectcache;
[16:53:20] 	Thanks.
[16:53:44] 	But you're still screwed, because you're got two wikis fighting over whose URLs to use
[16:53:55] 	Because they share parser caches
[16:54:21] 	I removed the other mediawiki software.
[16:56:05] 	Simetrical, anomie: Does PGSQL recognize STRAIGHT_JOIN? Or does it just choose the right order in a sane way?
[16:56:42] 	RoanKattouw, probably the abstraction layer will drop the STRAIGHT_JOIN, assuming you're passing it as an option to $dbr->select() and not manually concatenating bits of SQL?
[16:56:46] 	Thank you very much Roankattouw; I truncated the object cache, and my wiki seems to be back to normal :-O
[16:57:08] 	Simetrical: Yes, I'm doing it the right way, obviously. The abstraction layer inserts /*! STRAIGHT_JOIN */ somewhere
[16:57:21] 	Which MySQL does recognize, but I guess PG and others just see it as a comment
[16:57:22] 	The MySQL one does, pgsql likely doesn't.
[16:57:26] 	Or that, yeah.
[16:57:28] 	Whichever.
[16:57:41] 	Well then let's hope pgsql is sane enough to choose the right join order
[16:59:45] 	It likely will be.
[16:59:52] 	If not, that's G_SabinoMullane's problem.  :)
[16:59:59] 	Yeah
[17:00:58] 	Well my only remaining concern is is that MySQL 4 won't be smart enough to realize the GROUP BY is redundant after ns, title
[17:01:37] *Simetrical 	just wrote an actual, physical letter for the first time in, like . . . five years, probably?
[17:02:00] 	lol
[17:02:12] 	Odds are it's either a job application or a love letter :D
[17:02:26] 	That's pretty much the only thing people actually write these days
[17:02:47] 	Princeton is apparently unable to accept credit card payment if you've been denied a fee waiver for application to their graduate school.
[17:02:54] 	Well and little stuff like Post-Its and postcards and stuff, but that doesn't count
[17:03:00] 	They need a check mailed.
[17:03:06] 	Why, God only knows.
[17:03:10] 	Idiocy and incompetence.
[17:03:14] 	Checks? You guys actually *use* those?
[17:03:28] 	Of course, they didn't tell me the criteria for a fee waiver beforehand, so I had no idea if I would get it.
[17:03:30] 	I thought they would've been extinct by now
[17:03:32] 	Yeah, sure, all the time.
[17:03:34] 	What do you use?
[17:03:36] 	...
[17:03:44] 	How about bank cards?
[17:03:47] 	Cash is unsafe, not everyone accepts credit cards.
[17:04:01] 	(And by that I don't mean credit cards, but what in the UK would be called debit cards)
[17:04:02] 	Are those like credit/debit cards?  Debit cards are unsafe too.
[17:04:11] 	Not if they have a PIN on them
[17:04:17] 	Which is mandatory over here
[17:04:27] 	Well, they aren't so widely used here.
[17:04:37] 	03catrope * r44584 10/trunk/phase3/includes/api/ (ApiPageSet.php ApiQueryAllpages.php): (log message trimmed)
[17:04:37] 	API: (bug 16549) Kill a filesort in
[17:04:37] 	list=allpages&apfilterlanglinks=withlanglinks by replacing DISTINCT with GROUP
[17:04:37] 	BY. Of course GROUP BY page_namespace, page_title would suffice on MySQL, but
[17:04:37] 	Postgres will reportedly whine about that being against the SQL standard. Using
[17:04:40] 	GROUP BY on all selected fields instead (should please pgsql) and taking care
[17:04:42] 	that page_namespace, page_title is on front, which shouldn't cause a filesort is
[17:04:44] 	Pretty easy to get someone's PIN by watching over their shoulder or something.
[17:04:44] 	03(FIXED) list=allpages&apfilterlanglinks performance - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16549  +comment (10roan.kattouw)
[17:04:51] 	Then pickpocket the card.
[17:05:07] 	Yeah that happens sometimes
[17:05:09] 	With credit cards, you get the money back, the card company bears the loss.
[17:05:23] 	With checks I believe it's similar, it's their fault for accepting a bad check.
[17:05:55] 	With the cards over here, you're covered after you've called about losing your card, or if you can prove the transaction was fraudulent
[17:06:06] 	I don't know if that's true here.
[17:06:16] 	Anyway, plenty of places don't accept debit cards either.
[17:06:28] 	AFAIK those are generally just used for withdrawing cash.
[17:06:29] 	There was a fraud case recently where two Romanian guys got themselves locked in in a garden shop and replaced the card machine with one that copied all the cards and registered the PIN codes
[17:06:54] 	Cash, check, and credit card are the usual methods of payment.
[17:06:58] 	(here, that is)
[17:07:02] 	And maybe money order.
[17:07:06] 	They then used a copy of my mother's card to withdraw cash somewhere in Germany, while my mother had withdrawn cash a few minutes before hundreds of miles away
[17:08:11] 	She got a refund, of course, since it was quite clear the transaction was impossible
[17:08:29] *Simetrical 	is still leeching off his parents, doesn't need to care about this much :)
[17:08:36] *RoanKattouw 	too
[17:11:56] 	hrm... Will this filesort? http://rafb.net/p/PDalkq38.html
[17:16:02] 	This too: http://rafb.net/p/vNpp4Q29.html
[17:19:40] 	03(mod) clshow considers all categories !hidden - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=15579  (10b-jorsch)
[17:20:16] 	03(mod) clshow considers all categories !hidden - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=15579   +patch (10b-jorsch)
[17:29:28] 	Hi , I have one problem on Persian titles on my wiki .who can help me?
[17:31:05] 	hm... , any one there?!
[17:33:53] 	for example in this site, Farsi titles is broken : http://ln-s.net/2Zhs
[17:33:56] 	anomie: Are you the same person as b-jorsch on Bugzilla?
[17:34:08] 	RoanKattouw- yes
[17:34:17] 	anomie: Got commit access?
[17:34:22] 	nope
[17:34:25] 	Hmm
[17:34:31] 	OK, I'll commit your patch then
[17:34:41] 	thanks
[17:35:23] *anomie 	is working on a second patch to fix 'hidden' in list=allcategories and prop=categoryinfo, but needs to confirm no filesorting first
[17:35:44] 	no idea for this problem  ?
[17:35:45] 	anomie: OK. I'll be polishing your patch just a bit, though
[17:35:51] 	that's fine
[17:36:21] 	hamid_rostami: There seems to be no one around who knows anything about Persian letters or localization. Maybe you should try on a weekday
[17:38:42] 	root-80686: ok, tnx a lot :)
[17:39:37] 	RoanKattouw: :)
[17:40:04] 	03catrope * r44585 10/trunk/phase3/ (RELEASE-NOTES includes/api/ApiQueryCategories.php): API: (bug 15579) clshow considers all categories !hidden. Modified patch by Brad Jorsch
[17:40:12] 	03(FIXED) clshow considers all categories !hidden - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=15579  +comment (10roan.kattouw)
[17:40:57] 	does media wiki have any  email address for Persian support?
[17:41:41] 	ECA: Not that I know of. You could try contacting some people on http://translatewiki.org/ , those guys do all the translation stuff
[17:41:50] 	OK that URL's dead
[17:42:13] 	http://translatewiki.net/  that's the one
[17:42:59] 	RoanKattouw: thanks I will try this
[17:44:27] 	03catrope * r44586 10/trunk/phase3/includes/api/ApiQueryAllpages.php: Follow-up to r44584: finishing comment.
[17:44:47] 	03(mod) Feature: grant privileges to certain IP address ranges. - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=7492  +comment (10innocentkiller)
[17:48:21] 	tnx RoanKattouw for your help :)
[17:48:40] 	Simetrical: Sorry to be bugging you again :) but I've got another query here of which I don't understand why it filesorts
[17:49:08] 	http://rafb.net/p/GZKnJn30.html
[17:49:21] 	In my mind, this should be possible without filesorting:
[17:49:58] 	1) Retrieve the rows matching (pl_ns, pl_title) = (0, 'Foo') or (1, 'Bar') using the index on (pl_title, pl_ns, pl_from)
[17:50:13] 	2) Resolve the ORDER BY pl_ns, pl_title, pl_from using that same index
[17:50:50] 	3) Join on the page table and fetch the row (one, unique index) that matches pl_from=page_id using the PRIMARY index (page_id)
[17:51:05] 	Why doesn't it happen like that?
[17:52:25] 	03(mod) API protection query returning wrong result - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16629  (10public)
[17:52:27] 	RoanKattouw: ^
[17:53:03] 	MZMcBride: OK, those are definitely broken rows
[17:53:12] 	But they don't explain why the API lists them twice
[17:53:15] 	Can you delete them?
[17:53:35] 	Delete what?
[17:53:54] 	Those rows. Or is this a local DB dump copy?
[17:54:09] 	That's from the Toolserver, which is read-only. ;-)
[17:54:13] 	Right
[17:54:40] 	Maybe try to get an admin to delete and undelete the page
[17:55:10] 	Well, deleting them solves this particular issue, but what if there are other orphaned rows in the table? :-)
[17:55:29] 	They're not orphaned rows, they're corrupt rows
[17:56:04] 	MZMcBride: Can you find out how many of these rows there are? Like SELECT COUNT(*) FROM page_restrictions WHERE pr_level='';
[17:57:53] 	MZMcBride: One more thing, can you please run SELECT page_restrictions FROM page WHERE page_id=352339;  too ?
[17:58:04] 	I think that page may have leftover old-fashioned restrictions
[17:58:16] 	the query Special:WantedPages runs is pretty heavy and rather slow on my machine. i assume to speed this up would need additional caching/extra tables ?
[17:58:49] 	or does it get slower the more "wanted" pages there are. would it speed up if i made the most wanted pages? :)
[17:59:22] <^demon>	buzz_: all of the Wanted* queries suck, that's why they're cached on WMF :)
[17:59:38] 	^demon, so what can i do about it ? :)
[18:01:47] <^demon>	!misermode
[18:01:47] --mwbot--	I don't know anything about "misermode".
[18:01:50] <^demon>	!miser
[18:01:50] --mwbot--	I don't know anything about "miser".
[18:01:54] <^demon>	Stupid mwbot
[18:01:59] <^demon>	http://www.mediawiki.org/wiki/Manual:$wgMiserMode
[18:02:14] 	RoanKattouw- hrm, I can't say for sure, but one problem is probably that it's choosing the "pl_from,pl_namespace,pl_title" key instead of the "pl_namespace,pl_title,pl_from" key
[18:02:45] 	^demon, and how does mediawiki deal with the cache? it regenerated it occasionally?
[18:02:57] <^demon>	buzz_: You'll also need to  run ~/maintenance/updateSpecialPages.php whenever you want to update them.
[18:03:03] 	ahaa
[18:03:03] 	ok
[18:03:07] <^demon>	So set a cron to run it daily, weekly, whatever.
[18:03:26] 	anomie: Added FORCE INDEX(pl_namespace), but still filesorts
[18:03:38] 	RoanKattouw: http://paste.pocoo.org/show/95358/
[18:03:39] 	xchat died.
[18:03:41] *MZMcBride 	is away. :-)
[18:03:47] 	FYI, it's KEY (pl_namespace, pl_title, pl_from)
[18:03:58] 	MZMcBride: Ah, that explains. Thanks
[18:04:01] 	demon. ill run it daily i guess
[18:04:05] 	thanks!
[18:04:09] <^demon>	no problem.
[18:04:44] 	I'm trying to add a condidtion to the monobook title but fail to add a wg? gived me unexpected echo http://rafb.net/p/2mW1Is20.html
[18:04:58] 	gives
[18:05:15] 	Subfader: The wgNamespaceNumber and wgTitleF variables only exist in JS, not in PHP
[18:05:36] 	Also you need to drop the 'echo' there
[18:05:36] 	ok that's where i catched them :D
[18:06:07] 	You probably wanna use $wgTitle->getNamespace() == NS_CATEGORY
[18:06:21] 	And echo $wgTitle->getPrefixedText()
[18:06:44] 	And the unexpected echo is at if ( echo $GLOBALS['...
[18:06:48] 	(line 2)
[18:07:24] 	RoanKattouw- hrm... same explain output except "pl_namespace" in the key column?
[18:07:34] 	03(mod) API protection query returning wrong result - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16629  (10roan.kattouw)
[18:07:38] 	Yes. pl_namespace is the key it should use
[18:09:02] 	03(mod) API protection query returning wrong result - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16629  (10roan.kattouw)
[18:09:44] 	hmh. getNamespace() == NS_CATEGORY) { $this->text('pagetitle') ?>    Call to a member function getNamespace() on a non-object
[18:09:56] 	03(mod) Corrupt protections in page_restrictions table - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16629  summary; +comment (10roan.kattouw)
[18:10:00] 	Subfader: global $wgTitle; if (...
[18:10:26] 	anomie: Yes, same output. pl_namespace is the right key to use, so that's progress, but it still filesorts, which it shouldn't IMO
[18:11:29] 	RoanKattouw- hrm... Why TF is it using a full table scan over page instead of using the index?
[18:11:44] 	^demon, updateSpecialPages.php tells me "ALERT - script tried to increase memory_limit to 4294967295 bytes" (i think that is output from suhosin)
[18:11:49] 	That's what I'm trying to find out. Even FORCE INDEX(PRIMARY) on the page table doesn't take care of that
[18:11:53] 	seems a rather large number
[18:13:25] 	RoanKattouw- try SELECT STRAIGHT_JOIN ... FROM pagelinks, page ? I bet right now it's scanning page first and then pagelinks, despite the order of the explain output
[18:13:26] 	Simetrical: Any idea on http://rafb.net/p/5G7m5f72.html ?
[18:13:34] 	i assume it raises it so there is "no limit"
[18:13:37] 	anomie: Tried STRAIGHT_JOIN both ways, doesn't help
[18:14:01] <^demon>	buzz_: Dunno :-\
[18:14:08] 	One of the STRAIGHT_JOINs produces friendlier-looking EXPLAIN output, but still with 'using filesort'
[18:14:36] 	Wait a minute?
[18:15:09] 	Why TH is (pl_from, pl_namespace, pl_title) a UNIQUE index, while (pl_namespace, pl_title, pl_from) isn't? That's just crazy
[18:15:36] 	RoanKattouw- good question. File a bug?
[18:16:03] 	Yeah. Setting it to UNIQUE doesn't help though
[18:16:03] 	ok condition works now $this->text('pagetitle') gives me "Category:Page Name" but how do i get "Page Name" ?
[18:16:59] 	Subfader: What's $this in this context?
[18:17:09] 	in the structure of a mediawiki. is it a good idea to have a single top level category (called 'categories' or something) that all other categories live in ?
[18:17:25] 	buzz_: It could work. I know http://en.battlestarwiki.org/ does it
[18:17:36] 	There's no reason why it's a bad idea, anyway
[18:17:48] 	no idea. that's how monobook is written everywhere 
[18:18:01] 	RoanKattouw, i was thinking then i can have a easy link to browse all categories
[18:18:07] 	RoanKattouw, although there might be other ways
[18:18:25] 	buzz_: That's true, especially if you install CategoryTree
[18:18:50] 	buzz_: http://en.battlestarwiki.org/wiki/Special:CategoryTree/Category:A_to_Z
[18:19:20] 	yeh i like that
[18:21:27] 	03(NEW) Inconsistent indices on pagelinks, imagelinks, templatelinks tables - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16645 normal; normal; MediaWiki: Database; (roan.kattouw)
[18:22:13] 	^demon: Why don't you commit your patch to bug 7492? It looks quite simple
[18:22:39] <^demon>	RoanKattouw: Someone will complain that it doesn't belong in core :p
[18:22:43] <^demon>	!b 7492
[18:22:43] --mwbot--	https://bugzilla.wikimedia.org/show_bug.cgi?id=7492
[18:27:25] <^demon>	RoanKattouw: Opinion?
[18:34:40] 	i have more questions about categories. since there is an "uncategorised" files page. does this mean ideally all files should be categories ?
[18:34:51] 	same question for pages i guess too
[18:35:03] 	and templates ;-)
[18:35:34] <^demon>	That came from enwiki's obsessive desire to categorize _everything_
[18:35:58] <^demon>	And thus, it's Vitally Important that we know what isn't categorized :)
[18:36:12] 	I see
[18:36:48] 	i had a special page before (but needs to be updated for new layout), that showed pages which were not in any category and were not linked from any page (which means you could only get to them from the search).. might this be useful in core ?
[18:38:22] <^demon>	So an intersection of LonelyPages and UncategorizedPages?
[18:38:31] 	yeh
[18:40:10] 	my brain gets in trouble with categories. For example. say you have 10 pages each about an issue of a magazine called "gardener weekly". and you have a category called "gardener weekly magazines". . should you make a page with the overview of what gardener weekly is and link all issues in, or just put that on the category ?
[18:40:29] 	or both
[18:41:10] 	or am i misinterpreting when to use a category and when to just link the pages in.
[18:41:41] 	(this is not a wikipedia question, but one about the best ways of using mediawiki)
[18:41:44] 	03ialex * r44587 10/trunk/phase3/includes/specials/SpecialPreferences.php: 
[18:41:44] 	* removed useless check for $wgAllowUserSkin since we're already checked for it 18 lines above
[18:41:44] 	* wfHidden() -> Xml::hidden()
[18:42:40] 	hi all!
[18:43:07] 	I'm running mediawiki for an internal wiki on an Ubuntu server box.
[18:43:42] 	do you have an idea of the reason of this error?
[18:43:45] 	"Fatal error: Allowed memory size of 20971520 bytes exhausted (tried to allocate 22259 bytes) in /usr/share/mediawiki/languages/Language.php on line 1017"
[18:44:27] 	memory_limit in php.ini is 16M
[18:44:27] 	grisby, need to raise php limit
[18:44:35] 	make it 32mb or so
[18:44:35] 	I tried with 32 but no way...
[18:44:42] 	even more?...
[18:45:00] 	i have 64M i think
[18:45:35] 	grisby: 20971520 bytes is 20 Mo
[18:45:35] 	64M doesn't change... (I restarted apache)
[18:46:00] 	ialex: that's why I thought it wasn't memory_max but something else, actually...
[18:46:17] 	my php.ini is memory_max=16M (confirmed with phpinfo())
[18:46:28] 	memory_*limit*
[18:46:35] 	sorry.
[18:46:57] 	that's what I intended to say :)
[18:47:06] 	hehe :)
[18:47:52] 	grisby: do you have a line like "ini_set( 'memory_limit', '20M' );" in LocalSettings.php?
[18:48:42] 	Oh, Jes'!
[18:48:49] 	I had looked for set_ini...
[18:48:57] 	not ini_set ! :)))
[18:49:11] 	I definitely have a problem with my own memory...
[18:49:18] 	:D
[18:49:52] 	thank you very much! it's running now ;)
[18:49:59] 	np :)
[18:59:59] 	hello
[19:01:31] 	what's the php pendant for wgTitleF (page name without ns prefix)?
[19:01:59] 	04(REOPENED) Time based anti-vandalism feature - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=4397  +comment (10JSchulz_4587)
[19:02:45] <_Brun0_>	does mediawiki use a custom session handler?
[19:03:44] <_Brun0_>	because i have a cms in http://localhost/cms/ and wikimedia is on http://localhost/wiki/ and session variables in /cms can't be seen in /wiki. please help i'm trying to do auto-authentication.
[19:10:54] 	why is it not possible to "move" files ? would that be so hard to implement ?
[19:11:16] 	without a redirect.. id just like to rename some stuff
[19:14:34] 	03demon * r44588 10/trunk/phase3/ (15 files in 5 dirs): War on wfElement() and friends. Call the Xml members directly, rather than using old wrappers.
[19:15:12] 	\o/
[19:15:33] <^demon>	That should be all of them :)
[19:16:03] 	\o/ ^2 :)
[19:23:07] 	$wgExtraSubtitle = "test"; if i enter this into LocalSettings.php where does the text appear? It seems not to be working or I can not find it.
[19:24:39] 	ok i'm too dum for the monobook code. i wan to replace the pagetitle in the title tag to a title without the namespace prefix but fail totally
[19:32:04] <^demon>	Subfader: Just the  part, or that and the <h1>?
[19:32:24] <Subfader>	just $this->text('pagetitle')
[19:33:26] <^demon>	Replace that (in the <title> tags) with "global $wgTitle; echo $wgTitle->getText();"
[19:33:31] <^demon>	Without quotes, obviously.
[19:35:19] <ialex>	H4U: only works for some skins (CologneBlue, Nostalgia and Standard)
[19:35:54] <ialex>	H4U: but you can edit MediaWiki:Tagline to have the same result for all skins
[19:35:54] <H4U>	Ahh. im useing deafult.
[19:36:39] <Subfader>	oh that was easy. echo hrr. thanks a bunch
[19:36:56] <^demon>	No problem.
[19:37:05] <wikibugs>	03(NEW) RSS should put edit summaries, not just pagenames, into < title>s - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16646 15enhancement; normal; MediaWiki: History/Diffs; (jidanni)
[19:37:40] <H4U>	ialex, Im new to this took me 30mins to figure out mediawiki does not have admincp.. lol
[19:37:50] <H4U>	What file do i change MediaWiki:Tagline in?
[19:38:06] <ialex>	H4U: the MediaWIki:Tagline page
[19:39:24] <H4U>	Thx
[19:41:47] <_Brun0_>	I've implemented AuthPlugin in my mediawiki 1.9.6 to authenticate users from an external database. Now I need users that are logged in my CMS to be automatically logged in the wiki also. Sort of a single-sign on. What should I do?
[19:44:31] <wikibugs>	03(mod) UTF-8 in commit message confuses CodeReview - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=16640  +comment (10roan.kattouw)
[19:44:48] <wikibugs>	03(mod) UTF-8 in commit message confuses CodeReview - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=16640  (10roan.kattouw)
[20:02:37] <H4U>	How do i change skin? i tried $wgDefaultSkin = "cologneblue"; in localsettings.php 
[20:02:51] <RoanKattouw>	H4U: You have to change it in your preferences as well
[20:03:11] <H4U>	i see. Thanks ronan.
[20:03:55] <H4U>	What an ugly skin. lmao
[20:05:30] <Splarka>	"That's what she said"
[20:07:56] <wikibugs>	03(NEW) Neither list=allcategories nor prop= categoryinfo actually return the 'hidden' property - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16647 normal; normal; MediaWiki: API; (b-jorsch)
[20:08:45] <RoanKattouw>	anomie: Did you verify that the queries in your patch don't filesort?
[20:09:53] <anomie>	I loaded about 200000 rows from enwiki's latest page.sql dump, along with all of the category.sql dump and all of the page_prop.sql dump, and the queries didn't filesort. I couldn't manage to load the entire page.sql dump, too big.
[20:10:19] <_Brun0_>	Can someone tell me why my UserLoadFromSession is never called?!
[20:10:26] <_Brun0_>	the hook*
[20:11:22] <RoanKattouw>	_Brun0_: It should be. Are you sure you're not doing something else wrong?
[20:11:23] <_Brun0_>	Or better, can someone tell me in which ocasion is the UserLoadFromSession hook called?
[20:12:13] <RoanKattouw>	From what I know, it's called when User.php tries to create a User object from session data
[20:12:27] <RoanKattouw>	i.e. when someone who's already logged in visits a page
[20:12:43] <Splarka>	RoanKattouw: is https://bugzilla.wikimedia.org/show_bug.cgi?id=16647 a dupe of https://bugzilla.wikimedia.org/show_bug.cgi?id=15579 ?
[20:12:56] <RoanKattouw>	Splarka: No. Same bug, different modules
[20:13:06] <_Brun0_>	RoanKattouw, I have this in the end of my LocalSettings.php: http://pastebin.com/d71a0de17
[20:13:09] <Splarka>	well, same cause, right?
[20:13:45] <RoanKattouw>	_Brun0_: Are you sure $wgAuth->UserLoadFromSession() is the right function to call?
[20:13:52] <RoanKattouw>	Splarka: No
[20:14:12] <Splarka>	ahh k
[20:14:24] <_Brun0_>	RoanKattouw, i created that function and for a test pourpose I placed: var_dump($_SESSION); exit; on the first line of that function.
[20:14:42] <_Brun0_>	RoanKattouw, still nothing happens.
[20:14:43] <RoanKattouw>	_Brun0_: Did you get the capitalization right and all that?
[20:16:05] <RoanKattouw>	anomie: Just being paranoid here: did you also check your patch works?
[20:16:13] <_Brun0_>	RoanKattouw, see: http://pastebin.com/d316b29b8. As you can see it is written correctly.
[20:16:42] <anomie>	RoanKattouw- worked for me on PG, feel free to double-check it yourself
[20:16:45] <RoanKattouw>	Hmm yeah you're right
[20:17:07] <wikibugs>	03(mod) Errors in centralnotice.js - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16641  (10tfinc)
[20:18:08] <Simetrical>	RoanKattouw, never, ever use OR in MySQL.  MySQL does not like OR.  :)  At least in 4.0, for sure.
[20:18:16] <Simetrical>	It tends to just give up and ignore indexes.
[20:18:20] <Simetrical>	Use a UNION instead if necessary.
[20:18:29] <Simetrical>	(that will require a filesort, but a small one)
[20:18:40] <RoanKattouw>	Simetrical: Ah OK. Hmm the fact that a UNION is better kind of surprises me
[20:18:43] <wikibugs>	03(mod) Some diffs are escaped thus appear in single line - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16643  (10JSchulz_4587)
[20:18:55] <RoanKattouw>	Simetrical: How small is that? As small as the result set is? What if it's 5000 rows?
[20:18:58] <Simetrical>	It kind of surprises anyone who's not mentally disturbed, but it's true.
[20:19:05] <Simetrical>	It will filesort the result set, basically.
[20:19:10] <RoanKattouw>	Right
[20:19:20] <RoanKattouw>	How bad is that if the result set is 5,000 entries?
[20:19:40] <Simetrical>	Since you have to do (SELECT ... ORDER BY x LIMIT y UNION SELECT ... ORDER BY x LIMIT y) ORDER BY x LIMIT y, that's filesorting 2y rows.
[20:20:04] <Vyznev>	Simetrical: MW does use OR in some core code, so it's presumably not always completely awful
[20:20:11] <Vyznev>	for example, FileBatch uses OR
[20:20:13] <Simetrical>	Vyznev, yeah, probably not *always*.
[20:20:17] <RoanKattouw>	Simetrical: No, I'd have to filesort ny rows because I've got n UNIONS where 1 < n <= 5000
[20:20:19] <Simetrical>	What query?
[20:20:42] <Simetrical>	RoanKattouw, . . . ah.  That might be an issue.  I think 5.0 or 5.1 should be better at this (automatically merge sorting).
[20:20:43] <_Brun0_>	RoanKattouw, please help. You can see in my AuthPlugin (http://pastebin.com/d316b29b8) that the hook function is written correctly. Still it's never called. I tested it outside the class still no success. WHat can it be?
[20:21:04] <Vyznev>	Simetrical: see FileBatch::constructSet()
[20:21:13] <RoanKattouw>	_Brun0_: I have absolutely no idea, I have no experience with AuthPlugins
[20:21:24] <CIA-8>	03ialex * r44589 10/trunk/extensions/TrustedXFF/TrustedXFF.php: Added link to http://www.mediawiki.org/wiki/Extension:TrustedXFF in extension's credits
[20:21:38] <wikibugs>	03(mod) Parser hook output block level corruption - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=8997  +comment (10innocentkiller)
[20:21:44] <_Brun0_>	RoanKattouw, thank you for your attention
[20:22:00] <_Brun0_>	Can any one help me. My UserLoadFromSession hook is never called!!!
[20:22:42] <RoanKattouw>	_Brun0_: Not to be discouraging or anything, but since it's a Sunday there are fewer devs around. You might wanna try again tomorrow
[20:23:11] <_Brun0_>	RoanKattouw, thanks
[20:23:53] <ialex>	_Brun0_: you have MW 1.9.6?
[20:24:04] <_Brun0_>	ialex, yes
[20:24:10] <RoanKattouw>	Ah
[20:24:19] <ialex>	_Brun0_: use http://www.mediawiki.org/wiki/Manual:Hooks/AutoAuthenticate
[20:24:37] <CIA-8>	03catrope * r44590 10/trunk/phase3/ (3 files in 2 dirs): API: (bug 16647) list=allcategories, prop=categories don't return "hidden" property for hidden categories. Patch by Brad Jorsch
[20:24:37] <_Brun0_>	ialex, i otught that was old. isn't it?
[20:24:39] <ialex>	UserLoadFromSession only exists since 1.3
[20:24:49] <ialex>	*1.13
[20:25:01] <^demon>	Today is API bug day :)
[20:25:03] <_Brun0_>	ialex, what's the latest stable?
[20:25:09] <ialex>	_Brun0_: 1.13.2
[20:25:12] <RoanKattouw>	^demon: Yep.
[20:25:17] <RoanKattouw>	Lot of stuff to catch up to
[20:25:41] <Simetrical>	Vyznev, do you mean LinkBatch?  I can't find FileBatch on trunk.
[20:25:52] <Splarka>	is Monday in Aus now? API Bug Monday?
[20:25:59] <_Brun0_>	sorry for taking you time RoanKattouw and ialex. My mistake.
[20:26:25] <^demon>	RoanKattouw: I wonder when we'll get our schema changes :)
[20:26:28] <Simetrical>	Mmm, yeah, it does use OR.  Interesting.  Well, it doubtless works in some cases.
[20:26:34] <Vyznev>	Simetrical: yeah, sorry, typo/thinko :)
[20:26:44] <Simetrical>	But it's something to watch out for.
[20:26:48] <RoanKattouw>	Splarka: I'm not in Aus :)
[20:27:11] <RoanKattouw>	But yeah, it's probably Monday there. It'll be Monday here soon as well
[20:27:19] <^demon>	7:27am Monday (EST) - Time in Sydney, New South Wales
[20:27:52] <_Brun0_>	here in Brazil, GMT-4 is 17:27 PM Sunday. =)
[20:28:29] <Splarka>	Bug Smunday
[20:28:36] <RoanKattouw>	^demon: Well Brion told me that was gonna happen when Tim finished this script of his to commit schema changes more efficiently
[20:28:39] <^demon>	You're two hours ahead of me, it's 15:28 Sunday :)
[20:28:47] <^demon>	RoanKattouw: Yes I know :)
[20:28:56] <RoanKattouw>	But I gotta poke Brion tomorrow anyway about diffs in prop=revisions
[20:29:03] <RoanKattouw>	So I'll ask him again
[20:29:10] <Vyznev>	Splarka: is bug Sunday where we make the bugs to fix tomorrow? ;)
[20:29:13] <RoanKattouw>	(and ask him to ask Tim to hurry up a bit)
[20:29:47] <wikibugs>	03(FIXED) Neither list=allcategories nor prop= categoryinfo actually return the 'hidden' property - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16647  +comment (10roan.kattouw)
[20:29:50] <RoanKattouw>	anomie: Thanks for the patches; I like it when people fix my bugs for me ;)
[20:30:11] <anomie>	RoanKattouw- no problem
[20:30:53] <RoanKattouw>	anomie: methinks you're the most active code contributor that doesn't have SVN access. Hell, you're more productive than some that do :P
[20:31:23] <anomie>	RoanKattouw- And I don't even contribute all that much, really
[20:32:35] <RoanKattouw>	Well you contribute to Bugzilla quite a bit too :P
[20:33:15] *anomie 	discovers the innodb_file_per_table option, which is apparently required to ever reclaim the disk space from the aborted attempt to import enwiki's page.sql dump. Fun.
[20:34:43] <RoanKattouw>	Typical. Some guy asks an API question 15 mins before I join, then leaves before I find out about it
[20:35:19] <^demon>	RoanKattouw: I was thinking about meta=siteinfo.
[20:35:28] <RoanKattouw>	What about it
[20:35:51] <^demon>	What about a param to include extension credits with siteinfo?
[20:36:10] <RoanKattouw>	Why'd you want those?
[20:36:35] <^demon>	Checking via API to see if a wiki has an extension installed?
[20:37:00] <RoanKattouw>	OK, yeah that's a good one
[20:37:14] <RoanKattouw>	<brokenRecord>Please file a bug about that</brokenRecord>
[20:38:19] *^demon 	goes and writes a patch first
[20:39:14] <RoanKattouw>	^demon: About bug 7492 (autopromote stuff), I'd say it's a nice feature, cleanly implemented and IMO you should just commit it and see what Brion does :P
[20:39:44] <RoanKattouw>	It's just a few lines, so poking Brion about it first isn't really worth the trouble
[20:40:09] <Reedy>	Why's the API not give "Media" as an alias for http://en.wikipedia.org/w/api.php?action=query&meta=siteinfo&siprop=namespaces|namespacealiases ?
[20:41:00] <RoanKattouw>	Reedy: Because Media: and File: are different namespaces. Look at the IDs for instance
[20:41:19] <Reedy>	oh, yeah
[20:41:21] <Reedy>	dug
[20:41:22] <Reedy>	duh
[20:41:31] <Reedy>	but its there
[20:41:33] *Reedy 	smacks his head
[20:43:31] <^demon>	RoanKattouw: Okay. Let me finish the extensions stuff first.
[20:43:45] <RoanKattouw>	Sure
[20:44:07] <RoanKattouw>	Feel free to commit that rightaway as well; if there's something wrong I'll fix it up
[20:46:20] <CIA-8>	03catrope * r44591 10/trunk/extensions/AdvancedSearch/ (10 files): 
[20:46:20] <CIA-8>	Adding AdvancedSearch extension to SVN per request on wikitech-l. Note that this
[20:46:26] <CIA-8>	extension is not meant for production use, but rather as a starting point to
[20:46:31] <CIA-8>	develop a decent category intersection framework, which means it'll be changed
[20:46:31] <CIA-8>	heavily in the time to come (I hope)
[20:49:33] <CIA-8>	03demon * r44592 10/trunk/phase3/ (RELEASE-NOTES includes/Autopromote.php includes/Defines.php): (bug 7492) Allow assignment of rights to specific IPs/ranges
[20:49:37] <wikibugs>	03(FIXED) Feature: grant privileges to certain IP address ranges. - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=7492  +comment (10innocentkiller)
[20:50:07] <CIA-8>	03aaron * r44593 10/trunk/extensions/FlaggedRevs/specialpages/RatingHistory_body.php: Make getUserList() group properly for logged in users
[20:52:24] <RoanKattouw>	^demon: Added your APCONDs at http://www.mediawiki.org/wiki/Manual:$wgAutopromote#Details
[20:52:53] <wikibugs>	03(NEW) FlaggedRevs #mw-data-after-content overflows quickbar w/ standard classic skin - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16648 major; normal; MediaWiki extensions: FlaggedRevs; (marcin.cieslak)
[20:54:39] <wikibugs>	03(mod) Add "diff to last sighted version" to watchlist - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=16636  15enhancement->normal; normal->high (10marcin.cieslak)
[20:56:17] <wikibugs>	03(mod) Add "diff to last sighted version" to watchlist - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=16636  normal->15enhancement; high->normal (10marcin.cieslak)
[20:58:22] <_Brun0_>	RoanKattouw, my UserLoadFromSession hook now works. But one problem remains, var_dump($_SESSION) in UserLoadFromSession() shows session variable as null! How can I access the session deom that hook?
[20:58:36] <_Brun0_>	from*
[20:59:41] <ialex>	call wfSetupSession();
[20:59:55] <ialex>	and then use $_SESSION
[21:00:43] <CIA-8>	03aaron * r44594 10/trunk/extensions/FlaggedRevs/specialpages/ReaderFeedback_body.php: Disallow reviews with no user ID or IP
[21:04:17] <^demon>	Superglobals are ew :(
[21:04:24] <^demon>	I rewrote WebRequest to not use them :)
[21:04:32] <_Brun0_>	ialex, thanks. but session is now empty array. Means no variable is set. Still, on the same server, in the cms. The user is logged in and a var_dump in it shows many variables related to user authentication. How can I see session variables from http://localhost/cms in http://localhost/wiki ?
[21:05:13] <ialex>	_Brun0_: you need to have the same session's name
[21:06:24] <_Brun0_>	ialex, you mean the PHPSESSID ? Or the session variable name?
[21:06:39] <ialex>	_Brun0_: PHPSESSID
[21:06:59] <ialex>	_Brun0_: you can change it in $wgSessionName if you need
[21:07:07] <ialex>	!wg SessionName
[21:07:07] --mwbot--	http://www.mediawiki.org/wiki/Manual:%24wgSessionName
[21:09:11] <wikibugs>	03(mod) Add "diff to last sighted version" to watchlist - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16636  (10JSchulz_4587)
[21:11:00] <^demon>	RoanKattouw: Thanks. Almost have the extensions working.
[21:14:17] <_Brun0_>	ialex, THANK YOU!!!
[21:14:34] <ialex>	:)
[21:15:14] <^demon>	RoanKattouw: Any ideas why this would break? http://pastey.net/104376 I'm getting an exception.
[21:15:52] <CIA-8>	03aaron * r44595 10/trunk/extensions/FlaggedRevs/specialpages/OldReviewedPages_body.php: Remove unused field
[21:16:10] <RoanKattouw>	^demon: What's the exception?
[21:16:22] <^demon>	Exception Caught: Internal error in ApiFormatXml::recXmlPrint: (author, ...) has integer keys without _element value. Use ApiResult::setIndexedTagName().
[21:16:55] <RoanKattouw>	^demon: Is $extensions even defined anywhere?
[21:17:01] <RoanKattouw>	Oh yes of course
[21:17:36] <^demon>	Ah, I think this is it. authors is potentially an array() of names. That would break it?
[21:17:56] <ialex>	yes
[21:18:04] <RoanKattouw>	^demon: Yes. setIndexedTagName() that too
[21:18:20] <RoanKattouw>	Or use setIndexedTagName_recursive() if that makes sense
[21:19:07] <^demon>	I did something easier :)
[21:19:25] <^demon>	$ext['author'] = implode( ', ', $ext['author'] );
[21:19:53] <ialex>	after a is_array() check? :)
[21:20:11] <^demon>	And isset().
[21:21:09] <RoanKattouw>	ialex: (array)$ext['author'] is good enough
[21:22:18] <^demon>	Alrighty, final version: http://pastey.net/104377
[21:23:51] <RoanKattouw>	Looks good
[21:25:35] <^demon>	Actually, I think I'm going to standardize output a bit more. Theoretically, anyone can put any stuff into $wgExtensionCredits. Need a uniform output, not jut a dump of the array :)
[21:26:43] <RoanKattouw>	^demon: You're not using arbitrary *keys* in your output, are you? That gets real messy
[21:26:56] <RoanKattouw>	Ah, I see you're not, good
[21:28:14] <RoanKattouw>	Simetrical: Got a recommendation for my OR vs. UNION problem?
[21:28:21] <wikibugs>	03(WONTFIX) Time based anti-vandalism feature - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=4397  +comment (10ian.woollard)
[21:31:34] <^demon>	RoanKattouw: A bit more. Now gives standard 5 fields: type, name, description, author(s), version
[21:31:47] <^demon>	http://pastey.net/104378
[21:32:08] <RoanKattouw>	Watching a movie, hold on
[21:32:37] <^demon>	meh, +1 http://pastey.net/104379
[21:34:54] <_Brun0_>	Why $user = User::newFromName('bruno'); returns a User object with id = 0. That is, it means that the user does not exist on the wiki database. But when I check there IS a bruno user there. Any hints?
[21:37:47] <Platonides>	try 'Bruno'
[21:38:17] *Bdka 	grins at ialex ;-)
[21:44:09] <RoanKattouw>	_Brun0_: Use $user->getId(); instead of $user->mId
[21:45:05] <RoanKattouw>	^demon: Are you aware of the fact that type, name and description aren't guaranteed to be set?
[21:45:14] <RoanKattouw>	Oh wait type is, of course
[21:45:19] <RoanKattouw>	But name and description aren't
[21:45:29] <RoanKattouw>	Also, do you know about descriptionmsg?
[21:45:53] <RoanKattouw>	Oh and your indentation seems messed up on lines 16-20
[21:48:12] <CIA-8>	03aaron * r44596 10/trunk/phase3/includes/specials/SpecialWatchlist.php: 
[21:48:14] <CIA-8>	* Rewrite query the new way. Should make bug 16636 easier.
[21:48:14] <CIA-8>	* Remove commented out code
[21:48:15] <CIA-8>	* Break long lines
[21:55:21] <_Brun0_>	Platonides, RoanKattouw, thank you! it's working! Single sign-on! Login in cms and you're logged in wiki also! Now where can I edit mediawiki logout code to clean up cms session also?
[21:55:33] <RoanKattouw>	_Brun0_: There should be a hook for that one as well
[21:55:38] <RoanKattouw>	IIRC
[21:55:44] <_Brun0_>	RoanKattouw, i'll search
[21:55:51] <ialex>	!hook UserLogoutComplete
[21:55:51] --mwbot--	http://www.mediawiki.org/wiki/Manual:Hooks/UserLogoutComplete
[21:56:09] <ialex>	or
[21:56:15] <ialex>	!hook UserLogout
[21:56:15] --mwbot--	http://www.mediawiki.org/wiki/Manual:Hooks/UserLogout
[21:56:27] <_Brun0_>	ty all.
[21:56:34] <RoanKattouw>	_Brun0_: You probably want the former, since that actually gives you a useful User object
[22:00:12] <^demon>	RoanKattouw: Descriptionmsg is useless here?
[22:00:21] <RoanKattouw>	Why?
[22:00:27] <RoanKattouw>	Oh right interface language xD
[22:00:48] <RoanKattouw>	Well returning the *name* of the message could be useful, the client could run it through meta=allmessages
[22:04:17] <^demon>	RoanKattouw: I wrapped name and description in isset()s as well. Would be stupid to be undefined, but _someone_ will manage to break it and report a bug :)
[22:04:26] <RoanKattouw>	Exactly
[22:05:00] <^demon>	Committing.
[22:06:45] <CIA-8>	03demon * r44597 10/trunk/phase3/ (RELEASE-NOTES includes/api/ApiQuerySiteinfo.php): Support extension information in meta=siteinfo.
[22:08:33] <^demon>	<ext type="media" name="OggHandler" description="Handler for Ogg Theora and Vorbis files, with JavaScript player." descriptionmsg="ogg-desc" author="Tim Starling" version="r42275" /> 
[22:08:42] <RoanKattouw>	Looking good
[22:09:31] <_Brun0_>	Logout hook worked. Thanks
[22:10:19] <wikibugs>	04(REOPENED) Can't add tags to revisions any more - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16381  +comment (10roan.kattouw)
[22:11:01] <_Brun0_>	Now i have to remove "create new user" form. Also, guest users should not edit or create content. Any fast hint where to look for those?
[22:11:24] <_Brun0_>	found http://www.mediawiki.org/wiki/Permissions
[22:13:41] <Hermanon>	hi,i've created a database.sql.gz using mwdumper .how should i import it to mediawiki on xampp ?
[22:19:54] <Hermanon>	hi,i've created a database.sql.gz using mwdumper .how should i import it to mediawiki on xampp please ?
[22:21:29] <CIA-8>	03raymond * r44598 10/trunk/extensions/Renameuser/SpecialRenameuser_body.php: Add a missing CSS class for consistency
[22:33:30] <siebrand>	any French speakers in here?
[22:33:58] <Mike_lifeguard>	I might be able to muster a few sentences :\
[22:34:03] <RoanKattouw>	siebrand: I think ialex speaks French
[22:34:09] <RoanKattouw>	Oh wait, he's not here
[22:34:15] <siebrand>	RoanKattouw: indeed :)
[22:34:15] <wikibugs>	03(mod) updateArticleCount.inc. php can be made significantly more efficient - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16632  (10sean)
[22:34:22] <siebrand>	going to ask in #wikipedia-fr
[22:37:23] *Vyznev 	used to speak some French ten years ago, but it's pretty rusty now
[22:40:00] <wikibugs>	03(NEW) Include a system message on a page when it is protected - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16649 15enhancement; normal; MediaWiki: Page protection; (happy_melon)
[22:40:52] <acidic32>	hi
[22:40:59] <acidic32>	how do i change skin for all users?
[22:41:04] <acidic32>	!skin
[22:41:04] --mwbot--	Overview: <http://www.mediawiki.org/wiki/Manual:Skins>. Skin usage: <http://www.mediawiki.org/wiki/Help:Skins>. Gallery of CSS styles: <http://meta.wikimedia.org/wiki/Gallery_of_user_styles>. Writing your own: <http://www.mediawiki.org/wiki/Manual:Skinning>
[22:41:45] <Splarka>	acidic32: $wgDefaultUserOptions
[22:42:10] <Splarka>	but people who already set it in their preferences won't have it changed
[22:42:24] <Splarka>	there is also $wgDefaultSkin
[22:42:42] <Splarka>	see http://www.mediawiki.org/wiki/Manual:$wgDefaultUserOptions and http://www.mediawiki.org/wiki/Manual:$wgDefaultSkin
[22:42:56] <Splarka>	and if you're really set on setting it for everyone, you can disable all other skins, prolly
[22:43:29] <acidic32>	how do i do that from the webpate?
[22:43:48] <Hermanon>	i've created a database.sql.gz using mwdumper how should i import it to mediawiki on xampp please ?
[22:44:10] <buzz_>	is there a script to regenerate the entire search index ?
[22:44:43] <Splarka>	!localsettings | acidic32
[22:44:43] --mwbot--	acidic32: All configuration is done in LocalSettings.php (near the end of the file). Editing other files means modifying the software. Default settings are not in LocalSettings.php, you can look in DefaultSettings.php. See <http://www.mediawiki.org/wiki/Manual:Configuring_MediaWiki>, <http://www.mediawiki.org/wiki/Manual:LocalSettings.php>, <http://www.mediawiki.org/wiki/Manual:Config_script>, and <http://noc.wikimedia.org/conf>
[22:45:12] <Skizzerz>	buzz_: rebuildtextindex.php
[22:46:19] <buzz_>	why didnt i see that
[22:46:21] <buzz_>	thanks
[22:46:21] <Skizzerz>	probably cuz it doesn't have the word "search" in it :P
[22:47:36] <buzz_>	hmm it doesnt generate the table i see if it doesnt exist. jsut that i realised that currently im not backing up the myisam table with the search index. because I dont want to lock the database for ages (im using --single-transaction with mysqldump)
[22:47:56] <buzz_>	but i guess i dont need to back up the search index. i can just recreate the table myself, and run that script
[22:48:19] <buzz_>	or i suppose i could back it up without locking
[22:49:15] <buzz_>	(is that unsafe?)
[22:51:45] <H4U_>	Is it possible to order categories in a custom order instead of alphabeticly?
[22:54:25] <Vyznev>	H4U_: yes, by specifying a category sort key on each page
[22:54:40] <Vyznev>	http://www.mediawiki.org/wiki/Help:Categories
[22:54:40] <Vyznev>	see also the {{DEFAULTSORT:}} parser function
[22:55:13] <H4U_>	:) Thanks Vyznev i wll look into the sort key
[23:00:59] <wikibugs>	14(DUP) Include a system message on a page when it is protected - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16649  +comment (10herd)
[23:01:05] <wikibugs>	03(mod)  Indicate that the page being viewed is protected - subtitle message? - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=10347  +comment (10herd)
[23:01:06] <Splarka>	ugh
[23:01:32] <Splarka>	that duping just emailed 15 people
[23:02:30] <wikibugs>	03(mod) Update to cite.php refs: ids fixed, footnotes, globbed links - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=16294  +comment (10ncw33)
[23:02:50] <Vyznev>	would you like some spam with your spam, spam, spam, spam, spam, baked beans and spam?
[23:03:22] *Splarka 	puts on viking helmet
[23:04:56] <Charitwo>	is it time to pillage?
[23:08:02] <Splarka>	burn their women, rape their loot, and pillage their crops
[23:12:17] <Vyznev>	no, no, no, you burn the loot, rape the crops and pillage the *women*, silly
[23:13:43] <Splarka>	right, I always get that confused
[23:18:13] <buzz_>	the fierce vikings donk look so tough now, with their ABBA, and cheap pine furniture stores!
[23:18:23] <buzz_>	donk=dont..
[23:18:33] <buzz_>	k isnt even close to t.. i must be an idiot
[23:19:30] <Splarka>	the fierce ones all went to Greenland to die out
[23:26:45] <techman224>	Hey, is there going to be 1.14 soon?
[23:27:03] <techman224>	It's been almost four months
[23:28:33] <buzz_>	1.14 would be nice
[23:28:41] <buzz_>	you gotta send brion a present though
[23:29:50] <Splarka>	nah, send brion a past, he needs more retroactive time
[23:44:15] <Hermanon1>	 How to import a database in sql.gz format to mediawiki ,am using phpmyadmin ?
[23:52:38] <Hermanon1>	 How to import a database in sql.gz format to mediawiki ,am using phpmyadmin ?