[00:00:39] Tim: (briefly, hypothetically), what do you think about LOGO, ACTIONS, PERSONAL in Sidebar, to rearrange for non-floating template skins? (should have no effect in monobook, as they float, but Myskin for example) [00:04:34] why [00:04:37] ? [00:05:29] are you saying there's some CSS overlay for myskin where it actually matters? [00:06:10] nope, but it seems hackish to just do part of it [00:06:23] MediaWiki:BottomTwoThirdsOfTheSideBar [00:07:05] cactions and personal aren't in the sidebar [00:07:28] they can be with a little CSS (and are in Column one) [00:07:59] right, but it's not called [[MediaWiki:Column-one]] is it? [00:08:14] just sayin'! [00:08:27] and changing the position of the logo in that way wouldn't do anything [00:08:33] so it'd be a confusing feature [00:11:18] why not call it [[MediaWiki:NavigationBar]], the sidebar not be on the side either [00:11:48] err may not be [00:12:37] when is it not on the side? [00:12:37] when someone turns their CSS off? [00:14:44] <^demon> MediaWiki:Baroflinks :-) [00:14:47] we name it after the way it visually appears to most people [00:15:04] http://en.wikipedia.org/w/index.php?title=Main_Page&useskin=chick [00:15:36] http://en.wikipedia.org/w/index.php?title=Main_Page&useskin=simple [00:15:40] technical terms are hard enough as it is, especially when you consider everything to be pure data [00:16:16] we choose visual analogues to underlying concepts, to make them more memorable [00:16:28] especially for visually-oriented learners [00:17:14] it's the same reason I promoted the use of "red link" instead of "broken link" in core code [00:17:35] because even if your links are not red, you've probably seen them being red, and you know what a red link is [00:17:37] well, the obvious problem is, not everyone sees it the same [00:17:45] heh [00:18:04] so how come you dislike the description "Magic Word", they aren't magic, but they visually appear as magic to most people [00:18:20] *Splarka just playing devil's advocate there, bleh [00:18:58] magic like harry potter? [00:19:08] hocus pocus, pull a rabbit out of a hat? [00:19:11] on the latin wikipedia, probably [00:20:40] (oop, I guess they aren't localized, ahh well) [00:21:37] hehe, {{PAGINIS MOMINALE}}! [00:21:42] er N^ [00:24:08] Calling them red links in the core software assumes they'll always be red though. [00:24:34] There's nothing to say that the blue / red color-scheme won't change. [00:25:21] #CC2200 in gen=css, color: #ba0000; in monobook/main.css [00:26:03] I hereby declare that they will always be red [00:26:11] there we go, fixed [00:27:32] anyway, if we wanted to change the colour, we could just change the code [00:27:41] mmm, well, can you make them red then? for logged in users they are closest to "FireBrick" at 178,34,34 [00:28:07] (being 204,34,0) [00:28:10] that sounds like a clever name for red [00:29:03] if most people have seen a fire brick, I suppose ^_^ [00:29:28] http://en.wikipedia.org/wiki/Red [00:29:48] "Red is any of a number of similar colors evoked by light consisting predominantly of the longest wavelengths of light discernible by the human eye, in the wavelength range of roughly 625–740 nm." [00:29:55] but it still seems silly to have them a different color in gen=css for no apparent reason [00:29:58] not "Red is #FF0000" [00:30:09] *MZMcBride prefers #CC0000. [00:30:18] omg citing wikipedia [00:30:30] *GreenReaper prefers #204020 [00:30:39] Tim: Did you see the bug about global sessions not working correctly? [00:30:58] is it categorised under CentralAuth? [00:31:26] Oop. No, I put it in MediaWiki --> User login. [00:31:32] http://en.wikipedia.org/wiki/Red#In_popular_culture * Darth Maul from Star Wars: The Phantom Menace, had reddish/black skin, the makeup was achieved using Kryolan alcohol-based makeup to allow the actor to sweat without running [00:32:25] \n* Also in the Starr Wars universe, the light sabers used by most Sith were red, being made from artificially created emitter crystals. [00:32:30] *Splarka giggles at XKCD again [00:32:40] I don't see a CentralAuth component. [00:33:08] change the product first, and then it lets you choose a new component on the next screen [00:33:22] it doesn't do ajax just yet [00:33:43] There's still no CentralAuth component that I can see. [00:34:12] https://bugzilla.wikimedia.org/describecomponents.cgi?product=MediaWiki extensions [00:34:57] Oh, I switched to Wikimedia. [00:35:00] I should file a firefox bug about that [00:35:01] *MZMcBride growls at Bugzilla. [00:35:21] There we go. [00:35:25] (use jira) [00:35:25] 03(mod) Global sessions not working across domains - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14736 (10N/A) [00:35:27] hmmm, maybe I can just change it in my prefs [00:35:52] hmm for visual-oriented learners, so call it [[MediaWiki:SiteLinks]] [00:36:11] 'Cause that's not ambiguous... [00:36:55] Anyway, it seems global sessions are only per domain (*.wikipedia.org, *.wiktionary.org) like it says they are. [00:37:14] Or says they aren't. [00:42:25] MZMcBride: What's wrong with global sessions? [00:42:36] They don't seem to work. [00:42:49] I can log in to a Wikipedia and be logged into all Wikipedias. [00:43:08] But it doesn't log me in to any Wiktionaries or Wikiversities, etc. [00:43:15] I bet it's your third-party cookie settings. [00:43:49] lol " I hereby declare that they will always be red" [00:44:01] great problem-solving [00:44:38] Werdna: I reset my Safari (a full reset) and I'm having the same issue consistently. [00:44:43] I'm also having it in Firefox 3. [00:45:09] MZMcBride: Have you got your privacy settings set to reject third-party cookies without comment? [00:45:17] anyway, is there a guide to optimizing php.ini, apache, mysql for MW? (I have a my own VPS now, which means my own LAMP! :) ) [00:46:10] Werdna: Ahh. Well, that's a bit of a trade-off. [00:46:31] MZMcBride: It relies on your accepting third-party cookies. [00:46:51] Which advertisers also use. Can that particular note be mentioned somewhere? [00:47:50] Hmm. So the icon indicators seem to be misleading then. [00:48:17] I see all the icons, but unless I specify that Safari accept cookies from sites I haven't visited, it won't, regardless of icon presence. [00:53:35] hmm [00:53:52] Tim: Feasable to have a different image show if the cookie fails? [00:54:53] it ought to be, sites all the time tell you when you need to enable cookies [00:55:10] including mediawiki when you try to login [00:55:18] if you have cookies disabled completely [00:55:29] darkcode: yes, but these are simple [00:55:40] they aren't loading anything but an and setting a cookie [00:55:57] my question is: can the reply image be generated after the cookie is set and then tested... [00:55:57] (Or not setting a cookie, as the case may be.) [00:57:05] well the cookie can be set and tested for, before any type of response is shown [00:58:25] well, you could set it on a 302, then check for it on the second url [00:58:28] in theory [00:59:44] brion: sup [00:59:55] yo [01:02:50] 03(mod) Create a wap mobile gateway for the english wikinews similiar to wikipedia - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14755 (10brion) [01:07:13] hey brion [01:08:06] 03brion * r37360 10/trunk/phase3/ (RELEASE-NOTES includes/memcached-client.php): [01:08:06] Revert r37295 "(bug 14159) Use of MEMCACHE_COMPRESSED constant causes an E_NOTICE on PHP installs with the Memcache module installed. Renamed to MEMCACHE_MCOMPRESSED." [01:08:06] No such constant is modified here... there's a class constant instead which is being renamed, apparently for no reason. [01:10:50] 03(mod) Got notice, about MEMCACHE_COMPRESSED constant - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14159 (10brion) [01:17:55] 03brion * r37361 10/trunk/phase3/includes/specials/SpecialUserrights.php: Cleanup r37317 -- let's make this a little more legible. :) [01:21:07] What did I do now? [01:24:04] combining (bool)count and the ?: operator in one line gets you double points :D [01:24:39] What am I supposed to do? use TWO lines of code? [01:25:23] ;) [01:25:24] brion: Do you know if anyone has proposed openstreetmap integration in wikipedia? I dont see anything about it in bugzilla or the OSM mailing lists [01:26:14] avar: http://svn.wikimedia.org/viewvc/mediawiki/trunk/extensions/SlippyMap/ [01:26:16] *Werdna just does his part for keeping the download size of mediawiki down :p [01:26:29] <^demon> brion: I knew that was a class constant, but I couldn't find any other remotely possible explanations of the bug...thought maybe PHP was being odd and not liking ::COMPRESSED in associate with memcache. [01:26:38] Also, what the fuck is this: string(184) "" [01:26:46] ^demon: it was an old-fashioned define() in prior versions [01:26:50] how can you have an empty string which is 184 characters long. [01:26:51] the bug reports were against 1.12 [01:26:58] <^demon> Ah, that would explain why I didn't see it in trunk ;-) [01:27:07] Werdna: might be zero-width spaces ;) [01:27:14] brion: I used trim [01:27:21] trim might not get em [01:27:34] die( var_dump(trim($this->labeltext) ); [01:27:40] there should be a handy hexdump function... /me checks [01:27:56] bin2hex() looks handy [01:28:05] brion: JeLuF pushed it and you told him to implement caching?:) [01:28:17] caching would prolly be good :) [01:29:03] first few chars are 0a3c212d2d200a4e65775050206c696d6974207265706f72740a50726570726f636573736f7 [01:29:24] looks to me like garbage. [01:30:03] man, inputbox extension is a mess. [01:32:01] brion: I don't see anywhere that that particular var is even set! [01:32:08] but isset returns true [01:32:43] aha, tracked it down [01:33:00] <^demon> brion: I just looked at 1.12, and the constant is there. Mind if I backport a fix for it? [01:37:53] ^demon: no point to it imho [01:38:01] we won't get a 1.12.1 out before 1.13.0 i think :) [01:38:42] <^demon> True. But 1.12.1 might happen post 1.13.0 (security fix, etc added at some later point) [01:41:12] 03werdna * r37362 10/trunk/extensions/inputbox/inputbox.php: [01:41:12] Some fixes to allow inline search forms: [01:41:12] * Don't display an empty label if no label text is provided. [01:41:12] * Add an 'inline' parameter to the search form, which, if set, adds 'display: inline' to all the block elements, allowing it all to collapse' [01:42:41] eh, if you like then :D [01:43:08] no response on #firefox in irc.mozilla.org, nothing in the docs [01:43:26] in FF3, if you have %20 in a URL, it converts it to a space in the location bar [01:43:32] so when you paste into IRC, it's messed up [01:43:40] that's kinda annoying [01:44:07] it's a bit like the opera feature, which decodes unicode characters, but more broken [01:44:15] so I thought there might be a way to turn it off, but I can't find one [01:44:28] spaces in filenames and URLs, always bad, it will never be good [01:44:50] there'll never be a day when you can say "okay, everyone supports spaces perfectly in things that should be space-delimitable, problem solved" [01:45:28] you can't put a space in a URL, but you can put %20, in theory [01:45:35] http://ja.wikipedia.org/wiki/%E3%83%A1%E3%82%A4%E3%83%B3%E3%83%9A%E3%83%BC%E3%82%B8 hmmm [01:45:45] that's the idea of percent encoding, to avoid messing up the spaces [01:45:46] nothing wrong with displaying the decoded URL in the location bar, it should just re-encode it when you copy [01:45:49] it seems to be encoding the other chars [01:46:07] (at least on mac) [01:46:08] "but you can put %20" <- not with FF3 apparently, heh [01:46:21] *Splarka reiterates: will never not have problems somewhere [01:46:21] flyingparchment: if I suggested that to the mozilla developers, it would sound too much like work [01:46:28] they're still volunteers mostly, aren't they? [01:47:17] it's okay to to ask them to do hard stuff if they're paid? [01:47:19] i thought they inherited a whole load of Netscape developers [01:47:31] brion: that's when you use the clipboard [01:47:35] Splarka: never not? :) [01:47:48] Charitwo: yes [01:47:48] when you use the X register, you get http://ja.wikipedia.org/wiki/メインページ [01:48:14] cuts off after wiki/ for me [01:48:31] *Splarka gets � � � � � � � � � � � � � � � � � � (but mIRC sucks) [01:48:32] the plot thickens [01:48:46] iirc there's a filed gecko bug about nbsp and how it gets converted to a normal space when in the clipboard, wouldn't be surprised if it affected more then that [01:49:17] Splarka: yeah, so hard to find the configuration switch to enable its perfectly working UTF-8 support [01:49:18] nbsp gets converted to space in Firefox input forms too, no? [01:49:19] heh [01:49:42] OverlordQ: that's ancient, it's not related [01:49:43] sounds like a bug to me, it should be copying the same stuff either way on X11... [01:49:44] Tim: yes, especially in older versions where it doesn't exist [01:49:45] Splarka: that was supposely fixed [01:49:56] why would you use an old version? [01:50:02] you realise it's free, right? [01:50:04] omg I don't upgrade every day, I am a loser [01:50:17] you could upgrade once every couple of years [01:50:26] well, my firefox is 8 years old, and I don't have %20 changing to spaces [01:50:29] nyah nyah [01:50:34] >:D [01:50:40] it must have been at least 2 years since he added UTF-8 support to mIRC [01:50:59] oh, mirc finally supports utf-8? hah [01:51:10] yeah but like I say, it is actually hard to configure [01:51:21] it's a particularly obscure option, you'd have to know it was there [01:52:15] evil [01:52:23] why not support it out of the box? [01:52:51] like screen and irssi? [01:52:54] because it's mIRC? nothing make sense :) [01:53:06] you have to type screen -U instead of screen, and there's an obscure option in irssi for it [01:53:08] also it doesn't choose fonts depending on the characters it sees, like a browser does [01:53:21] on 1:MAKESENSE:*:fail() [01:53:25] so you have to select a font manually that has the characters you're interested in [01:53:35] fixedsys excelsior :) [01:53:50] http://www.w3.org/QA/2008/05/utf8-growth-google [01:54:24] Splarka: that must be all those mediawiki sites coming online [01:54:27] ;) [01:54:35] well, no [01:54:42] it must be all the forks and mirrors of wikipedia [01:54:57] the live ones [01:55:10] How does one graph Unicode usage? [01:55:26] check on web pages [01:55:30] or such? [01:56:11] I suppose. That sounds like a better test of tag usage, though... [01:57:07] they probably check both Content-Type header and the meta tag [01:57:14] send a HEAD and check the response Content-Type for charset? [02:02:02] https://bugzilla.mozilla.org/show_bug.cgi?id=440075 [02:02:18] it's no easy task, finding bugs in the mozilla bug tracker [02:02:40] it beats the php bug tracker [02:02:45] its no easy task, finding bugs in any bug tracker [02:02:52] any search i do there turns up 500 unrelated things marked as BOGUS [02:03:27] mozilla i usually just have to follow the trail of dupes with slightly different wording [02:03:42] the tricky thing about the PHP bug tracker is that the BOGUS bugs are often real bugs, that they just didn't feel like fixing at the time [02:03:56] the "no feedback" category in PHP is just as bad [02:03:59] full of real bugs [02:04:04] :( [02:04:13] "didn't bother to try reproducing, CLOSING" [02:05:00] there should be an IGNORE category [02:05:35] "closed I HATE YOU" [02:07:45] Werdna: re the inputbox commit [02:08:13] the inbputbox code is imho pretty terrible to begin with; it has a horrible tendency to take variables and change their format from some raw input to html-escaped output [02:08:34] the problem is i think some of the new code uses values before they've been escaped and puts them into output [02:09:33] *Werdna slaps Erik. [02:09:38] at least $id is getting used before it's run through Sanitizer::escapeId() [02:09:43] it's old code, it needs a rewrite :) [02:10:19] *Werdna shrinks away. [02:10:30] I tested it, and it seemed to come out alright. Did I mess anything up? [02:11:13] is it possible to change the search feature to be case in-sensitive? [02:11:26] i think just move the $id = Sanitizer::escapeId( $this->id ); line up should be fine [02:11:35] uberfuzzy: TitleKey extension [02:12:01] (the full-text search is case-insensitive already. TitleKey makes the 'go' exact-match checks case-insensitive as well) [02:12:33] gotcha, thanks [02:12:53] !e TitleKey [02:12:53] --mwbot-- http://www.mediawiki.org/wiki/Extension:TitleKey [02:15:02] TimStarling: What do we do about unattached local accounts which have no password because of that CentralAuth bug? [02:15:47] I'll reset their passwords manually [02:15:59] I've got them on ro and pt, I think [02:16:01] let me check [02:16:41] yes, ro and pt. [02:16:43] before I forget again... [02:17:04] brion: WAP gateway, google, I'm leaning towards "yes" [02:17:11] I'll send them an email now [02:17:17] sure [02:17:29] to be honest, i use it all the time on my iphone now ;) [02:17:33] it "mostly works" [02:17:49] do we want to consider renaming the url though? [02:17:57] wap is so.... lame :D [02:18:07] brion: Is it any good? I'm getting an iPhone in a few days :) [02:18:55] http://en.wap.wikipedia.org/transcode.php?go=hells+yeah [02:19:41] That's pretty cool. [02:20:00] people complain that it's a language code [02:20:08] heh [02:20:09] which is probably a fair complaint [02:20:20] en.mobile.wikipedia.org might be nicer or something [02:20:22] *brion-away back in a bit [02:20:29] yeah [02:20:56] well, most phones don't use WAP anymore, AFAIK [02:34:14] !rar [02:34:14] --mwbot-- rar [02:36:58] !rar del [02:36:58] --mwbot-- Successfully removed keyword: rar [02:42:02] Is there a nice way to get the full URL not including the article name, e.g. 'http://en.wikipedia.org/wiki/'? [02:46:09] nm, it looks like this does the trick: [02:46:12] $wgServer . str_replace( '$1', '', $wgArticlePath ); [02:53:37] 03(mod) Semantic Forms short-circuts the new wfLoadExtensionMessages system and tests incorrectly - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14357 (10yaron57) [03:04:03] 03brion * r37363 10/trunk/extensions/inputbox/inputbox.php: move some escaping up to before $id is used [03:05:11] brion: Oops, I forgot to commit that. [03:06:16] grr silly message updates [03:06:58] makes reverse merges take *forever* [03:09:21] I'm having an issue with an upgrade from Mediawiki1.7 to 1.12... It now lists an error saying a table that does exist does NOT. Any suggestions? [03:09:38] 03werdna * r37364 10/trunk/extensions/AbuseFilter/install.php: Take over accounts which already exist, so we're not giving random people admin access. [03:09:43] 'wikidb.w_page_restrictions' doesn't exist (localhost)". [03:09:58] ok, I've set up *.wap.wikipedia.org to redirect to *.mobile.wikipedia.org, and mobile.wikipedia.org redirects to en.mobile.wikipedia.org [03:10:17] Zaerath: can you double-check the table name prefix there? [03:10:22] assuming you're not behind the telstra DNS cache that I poisoned by negative caching it [03:10:40] my dns ain't seeing it either :( [03:10:43] brion: Yes, and it's correct. I used w, so it used w_table name. [03:10:53] hmm, i see en.mobile.wikipedia.org but not mobile.wikipedia.org [03:11:10] well, I checked directly via ns*.wikimedia.org and it seemed to work [03:11:13] brion: Wait; but now the table does not show. I used the installer; shouldn't it have created the table? [03:11:19] maybe it's cached somewhere in the servers as well [03:11:30] Zaerath: confirm the output from the installer [03:11:38] *.mobile.wikipedia.org works just fine, I didn't request that one early [03:11:40] did it abort before completing that upate? [03:11:44] brion: I think it may be a simpler issue than I originally saw. [03:12:01] [brion@zwinger php-1.5]$ host mobile.wikipedia.org [03:12:01] Host mobile.wikipedia.org not found: 3(NXDOMAIN) [03:12:24] I know, I think that's the resolver cache for the whole pmtpa cluster [03:12:38] hmm, something breaking on test [03:12:52] try dig @ns0 mobile.wikipedia.org [03:13:26] anyway, I'll wait for the caches to expire, and then send google an email, and announce it on wikitech-l [03:13:37] brion: Stupid syntax error on my part. Thanks a ton. [03:13:43] how is google involved? [03:13:54] they're asking if they should send tons of people to it [03:13:59] actually that's a point... [03:14:04] I hope it doesn't break under the load [03:14:30] heh [03:14:34] it might, it's not super-optimized [03:14:41] well, only one way to find out ;) [03:14:42] single server [03:14:45] lol [03:15:01] Jul 9 03:14:57 srv35 httpd[2415]: PHP Fatal error: Cannot redeclare wfmsg() (previously declared in /home/wikipedia/common/php-1.5/includes/GlobalFunctions.php:337) in /home/wikipedia/common/php-1.5/includes/MessageFunctions.php on line 44 [03:15:18] ah, conflict in GlobalFunctions [03:15:19] sigh [03:15:24] it's hard to see one C in a giant list :P [03:16:11] well, i'll just revert GlobalSettings [03:16:15] if it was important it'd have been committed [03:16:19] "svn status" is one way to work around that [03:16:27] it shows which files are conflicted [03:16:35] yes, and 500 other things :D [03:16:47] | grep ^C [03:16:54] yeah i'm just lazy [03:17:07] ok, scap time [03:17:14] yay! [03:17:22] brb [03:17:53] Odd :-/ [03:17:54] Notice: Undefined offset: -1 in /home/wiki/languages/Language.php on line 422 [03:34:07] Werdna: Poor grammar is poor grammar. http://en.wikipedia.org/w/index.php?title=Wikipedia:Abuse_filter&curid=18088626&diff=224493603&oldid=224485448 :P [03:46:21] Tag extensions get processed in Sanitizer::removeHTMLtags(), do they? [03:51:49] something change in the date stuff recently? [03:55:08] davidmccabe: nope [03:55:19] OverlordQ: why do you wask? [03:55:21] *ask [03:56:01] that php notice I pasted a bit ago, getting it with the latest svn revision [03:56:26] which language? [03:56:52] En [03:57:03] got a sample page that triggers it? [03:57:08] mainpage [03:57:14] brion: I need to place a marker in the text that tells me to begin and end a certain kind of processing. This processing concerns links. [03:57:38] http://www.thedarkcitadel.com/wiki/Main_Page [03:57:43] brion: at what part of the process should I hook in and add my markers to the next? [03:57:52] davidmccabe: Is this in LiquidThreads. [03:57:59] Werdna: no. [03:58:27] brion: I had preferred to use a tag extension so that I wouldn't have to worry abut NOWIKI or such. However that is a separate processing step. [03:58:34] Does my question even make sense to you? [03:59:14] hmmm [03:59:37] well, you can break the whole section out as a replacement and do sub-parsing (eg, wrapping the whole thing as a tag) [03:59:51] otherwise we don't really have a clean way of doing it other than horrible regex fun :D [04:00:12] Ah, so I would have everything inside a tag instead of begin and end tags? [04:00:21] may be possible to use the placeholders to stick in your start/end points and process those in a regex after they're replaced back [04:00:26] but that might be fragile too [04:00:30] Do you have handy what I need to do to cleanly invoke subparsing, or should I just look aroundud and figure it out? [04:00:39] It does exist. [04:00:50] done in the inputbox extension. [04:01:03] !inputbox [04:01:03] --mwbot-- http://www.mediawiki.org/wiki/Extension:Inputbox [04:01:11] hey hey, it worked. [04:01:16] $output = $this->parser->parse( $this->labeltext, [04:01:16] $this->parser->getTitle(), $this->parser->getOptions(), false, false ); [04:01:21] $this->labeltext = $output->getText(); [04:01:35] iw ouldn't recommend using inputbox as an example of anything, personally ;) [04:01:43] haha [04:01:45] I know [04:01:48] FYI, the contents of the tag is going to be most of the page. [04:01:48] it should use all the Xml methods [04:01:50] *Werdna hides. [04:02:01] So it has to be quite stable. [04:02:25] brion: I keep expecting everything to happen from top to bottom, and the fact that it doesn't is screwing with my brain. [04:02:54] OverlordQ: hmm, now i'm not seeing it anymore [04:02:56] I have gone through about three different solution-ideas that were all based on the premise of being able to change my own state as the parser goes through the page sequentially. [04:03:15] brion: yea I specific an error reporting function [04:03:19] *specified [04:03:23] ah [04:03:41] 73 meg log now :-/ [04:04:37] brion: so do you recommend the subparse approach you just mentioned or something else? [04:05:03] generally i'd probably do the subparse, but that may or may not be suitable for your requirements [04:05:13] In what case might it not be suitable? [04:08:53] strange, dont get set to the log [04:09:22] who knows :) [04:09:43] but currently anything inside a tag doesn't get {{subst}} and other post-save processing done :( [04:10:09] *Werdna bored, tempted to start work on another project but too lazy [04:12:00] only thing logged were notices about date/strtotime and TZ (aside from a few unrelated warnings) [04:12:59] anyone know of a release date for mediawiki 1.13? [04:13:23] ask brion :) [04:13:45] brion: Now would be a good time, if you have time to go through global blocking, as I'm here and have time to make any changes. :) [04:19:12] it's not a huge extension, and I fixed up all the things you told me to fix up when you reviewed it a while ago. [04:20:46] !wg Localtimezone [04:20:46] --mwbot-- http://www.mediawiki.org/wiki/Manual:%24wgLocaltimezone [04:31:21] Aahhhhh, I might have found it [04:37:01] brion: so I tried the subparse approach. Everything seems to work great... except there are weird changes to the output. [04:37:11] For example, a certain list item becomes indented. [04:40:53] :( [04:40:57] Werdna: it's on my list :D [04:41:04] i know i've been behind :( [04:41:43] brion: I don't mean to nag. I just thought now would be a good time, since I'm around and willing to make some changes. [04:42:03] dinner's on the table, sorry [04:42:12] gotta run or my lady will be very annoyed with me :) [04:42:15] late dinner [04:42:19] enjoy [04:42:19] ya [04:42:21] tx [04:43:04] 03(mod) Double accesskeys cause problems - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14757 (10brion) [04:44:40] awww [04:45:01] bad brion :P right when I find the problem [04:45:11] what's up, OverlordQ ? [04:45:50] http://pastebin.ca/1066325 [04:46:15] spitting out that PHP error (undefined index -1) when viewing main page [04:46:20] with latest svn rev [04:47:00] yeah [04:47:03] did you find the bug? [04:49:08] seems the substr is off by one [04:49:35] can you make a patch? :) [04:50:53] well if I knew why the problem started showing up now lol. Since browsing svn I dont notice anything related that's changed. [05:04:48] hopefully the right version :/ [05:18:39] yea I really dont understand it lol :( [05:22:06] 03dale * r37365 10/branches/MetavidWiki-exp/ (16 files in 9 dirs): [05:22:06] * search digest table updates (be sure to run mv_update.php) [05:22:06] * #mvData:TOPSEARCHES & #mvData:POPULARCLIPS magic Functions active with arbitrary granularity for view count summaries. [05:22:06] * style and layout fixes. [05:22:06] * clip view counts per mvd_page in search results [05:33:31] Ahhhhhhh, think I found it . . . maybe again [05:38:08] 03werdna * r37366 10/trunk/extensions/inputbox/inputbox.php: Oops. Fix weirdness [05:41:16] 03(NEW) PHP Strict Standards Issue in DatabasePostgres.php - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14763 15enhancement; normal; MediaWiki: Database; (overlordq) [05:41:19] 03(mod) PostgreSQL support (tracking) - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=384 (10overlordq) [05:50:17]