[00:06:54] at least as interoperable as posix ;) [00:07:03] which is to say, if you try to be portable it usually works [00:07:26] of course the standard doesn't cover things like the windows forms ui libraries -- that sort of area tends to be the scariest [00:07:35] or weird hybrid code that calls lots of windows native functions ;) [00:07:47] But neither does, say, the C standard. [00:07:49] Oh well. [00:07:51] yep [00:08:12] but you can write portable code pretty well -- there's a nice gtk stack that works on both unix and windows ;) [00:08:29] and i guess their windows.forms implementation mostly works but i've never had any need to work with stuff written for windows front-ends [00:08:51] just played with some server-side things and some client software written for the gtk stack (gnome stuff) [00:15:31] Man, why do we have agreement in C on how to do everything except OS services that go beyond command-line printing and basic file use? [00:15:36] :( [00:16:29] 03(FIXED) A new logo for Finnish Wikisource - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14861 +comment (10brion) [00:17:53] 03(mod) Add hook to add HTML at end of Email section of user preferences. - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14864 (10brion) [00:17:58] congrats brion, heard you were engaged [00:18:01] :) [00:19:11] *AaronSchulz watches the cat run out of the bag [00:19:16] 04(REOPENED) DifferenceEngine grinds on pathological case - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14867 +comment (10brion) [00:19:21] tx :) [00:23:02] brion: we are going to need photos you know :D [00:23:14] brion left [00:23:22] :( [00:23:30] *AaronSchulz kicks conference mode [00:24:03] *Alexfusco5 tells AaronSchulz to set conference mode to 0 [00:24:09] uh, probably just me being a complete noob, but I'm trying to set up a copy of MediaWiki 1.12.0 on my Mac, and I have no clue how to set up MySQL the right way [00:24:26] *AaronSchulz tells Alexfusco5 that that's too much work [00:25:09] can someone guide me through that part? [00:25:24] nihiltres, maybe you want #mysql? [00:25:28] heh, probably [00:25:49] I'll head there and come back if that doesn't help [00:26:46] *AaronSchulz awaits his return [00:26:57] heh [00:27:25] Soooo, good 'ole Simetrical, how goes it? [00:28:57] nihiltres: i think i just downloaded a mysql server for osx [00:29:11] pre-compiled, out of impatience [00:29:18] *Simetrical waves to AaronSchulz [00:29:30] well, yes, that's I think what I did [00:29:36] AaronSchulz, I'm busy shaking someone's faith in the objectivity of morality. [00:29:55] my real problem is configuring the user accounts so that MediaWiki can use it and there aren't gaping security holes [00:30:06] have fun with that :) [00:30:11] morality is subjective [00:30:26] exactly [00:30:28] and relative... you can't be right in "everything" [00:30:50] you can't put tigers on vegetarian diets ^_^ [00:30:59] Splarka: tell the former to Kant :p [00:31:05] can't [00:31:13] recant! [00:31:37] AaronSchulz, I think I've mostly convinced him. [00:32:43] morals are ethics enforced upon you [00:33:03] in the same way religions are philosophies forced on others [00:34:06] what is moral? primary conservation of self, of your offspring, of your life's work, of your family, of your community, state/country, religious or ethnic group, species, animal kingdom, all life, the surface of the planet, the whole planet, our solar system, the whole universe? [00:34:12] anyone have a clue why $wgOut->parse("{{".$wikitext."}}") wouldn't permit raw HTML (when $wgRawHTML = true)? [00:34:17] (back to mediawiki ;-) [00:34:17] which has precidence? [00:34:27] you're no fun ^_^ [00:37:16] my question always puts an end to all conversation :-( [00:38:35] are you sure that is the right parse function? the spaghetticode needs to be entered at the right point [00:38:55] well - it's the way Special:Call uses it [00:39:26] is the code called wrapped in ? [00:39:26] the preprocessor - which afaict normally deals with raw html - doesn't get called [00:39:32] wrapped, yes [00:39:50] is there some other way i could try calling parse() ? [00:39:53] is it... moral to do that? [00:40:16] *Splarka attempts conversation tangent again, which is moral, subjectively... and hides [00:43:28] don't ever do that AaronSchulz [00:43:40] why? :) [00:43:47] that is your only warning [00:44:06] oh, have a little humor [00:44:08] AaronSchulz: AGAIN! ;) [00:44:29] humor is one thing, kicking people from IRC channels for no reason is not humor [00:44:52] people, my parse problem still isn't solved, focus! ;-) [00:44:57] lol [00:45:04] Splarka: Maybe it would be funnier if he bacame a freenode staffer and used /kill ;) [00:45:27] Splarka: well the humor is just subjective, since I laughed [00:45:48] are you sure you wan't to adopt that attitude? [00:45:57] ^want [00:45:58] Splarka: and if the voice of all laugh's then we all laughed ;) [00:46:20] Splarka: bah, you're no fun anymore [00:46:55] *AaronSchulz gets his Game beta test key out [00:47:19] AaronSchulz: set mode +g so we can all invite spam ;) [01:01:57] 04(REOPENED) New revisions occasionally created with wrong text ( but correct rev_len) - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14933 +comment (10Platonides) [01:07:21] Simetrical: You about? [01:08:06] Splarka: but seriously, didn't mean to upset you :( [01:08:34] bah, conference mode again [01:09:57] YES! I SOLVED MY PROBLEM! [01:10:01] [01:11:01] MZMcBride, for a little longer. [01:11:28] __NOINDEX__ / __INDEX__ aren't supposed to work on ContentNamespaces by default, but I don't know what that means. [01:11:30] AaronSchulz, things don't work that way on this network. On some networks randomly kicking people is de rigeur, like QuakeNet, but not here. [01:11:35] Is that even-numbered namespaces? [01:11:41] MZMcBride, no. By default, it's the main namespace. [01:11:46] Ahh. [01:11:48] It's customizable. [01:12:01] On MediaWiki.org, it includes namespaces like Manual: and Help: [01:12:01] Silly name for a variable... [01:12:02] . [01:12:08] What is? [01:12:17] Well, I guess not a silly name if it's customizable. [01:12:29] A content namespace is one that contains content, as opposed to "meta" stuff about the project and its users. [01:12:33] Or discussions, etc. [01:12:53] I always thought of content namespaces as any non-talk namespace. [01:12:58] Nope. [01:13:01] But I see what you're saying. [01:13:16] Anyway, basically, if people deindex pages, it's not going to be anything we really need to be on Google anyway. :) [01:13:19] There's still no way to track the magic word usage, and it's going to spread rapidly soon... [01:13:21] Just somebody's user page or something. [01:13:35] So? Not like it's a big deal. [01:13:40] Or the Wikipedia: pages. :/ [01:13:43] It doesn't affect internal search, or important pages. [01:13:48] Sure. [01:14:00] "Haphazard application that's unmonitored" is my concern. [01:14:00] Some of those *shouldn't* be indexed. [01:14:14] Yeah, but everything in wikis is generally haphazard and unmonitored. [01:14:16] Of course. [01:14:26] This doesn't seem like a particular concern. [01:14:40] Of course it is. [01:14:48] If nobody cares enough about the page to notice random __NOINDEX__ creeping in, clearly it's not very important anyway. :) [01:14:49] What is unmonitored? [01:14:59] Categories, templates, images ... all montiored. [01:15:02] Monitored, too. [01:15:13] Pagelinks, transclusions, etc. --> all monitored. [01:15:17] Everything is unmonitored, unless you count recentchanges and similar things. Which track __INDEX__ too. [01:15:22] Every edit to every page is unmonitored. [01:15:32] Just as much as edits that add or remove __NOINDEX__. [01:15:34] Page protections have a Special: page to list them. [01:15:41] __TOC__ is unmonitored. [01:15:52] __TOC__ doesn't change Google's results. ;) [01:15:55] And you know that. [01:16:21] Adding "MZMcBride enjoys the company of female poodles" changes Google's results. [01:16:23] And do we really want people making thousands of edits in a mis-guided effort to noindex an entire namespace? [01:16:27] It adds that string to the page. [01:16:42] If people make thousands of stupid edits, that's their waste of time. [01:17:32] Basically, __NOINDEX__ is no worse than some random bit of libel, which is tracked no better. [01:17:44] Bah! [01:17:49] Maybe we could add a message to the page when you're editing it saying that the robot policy has been overridden. [01:17:56] We're (supposed to be) implementing flagged revisions for that. [01:18:06] Flagged revisions should work for this too. [01:18:34] "Their waste of time" and a giant waste of resources. [01:18:51] Why is there NO documentation for BoardVote? http://www.mediawiki.org/wiki/Extension:BoardVote [01:19:05] I mean what configuration options are there, or how do I enable it, etc.? [01:19:16] Oh, come on. Like we should avoid features because people will waste some of their time if they're too stupid to realize how to use them correctly? [01:19:19] That's an untenable attitude. [01:19:31] lymeca, because it's meant to be a Wikimedia-only extension? [01:19:41] We keep almost everything trackable. [01:19:47] lymeca, it's designed specifically for voting for the Wikimedia Board of Trustees on Wikimedia sites. [01:19:52] MZMcBride, this is every bit as trackable as any edit. [01:19:56] From edits to template usage to image usage to userrights changes. [01:19:59] oh [01:20:11] a noindex option for FlaggedRevs would be nice, and __NOINDEX__ seems pointless when anyone could just remove it, something like MediaWiki:no-index-pages would make more sense... [01:20:16] Simetrical: It could easily be more trackable. [01:20:24] MZMcBride, "easily"? Then submit a patch. [01:20:28] Heh. [01:20:30] ;) [01:20:54] I'm saying that the framework to track magic words and other weird parts of page text like this is already written and available. [01:21:01] darkcode, it's not pointless. It's meant for things like, say, RFA. Put it in the header template and you noindex all of them without having to constantly update robots.txt. [01:21:24] MZMcBride, what "framework to track magic words and other weird parts of page text"? [01:21:25] But we don't know if someone's done that already or not. [01:21:44] *AaronSchulz heard "FlaggedRevs" [01:21:45] darkcode, MediaWiki:No-index-pages doesn't scale to 2,000 back issues of AN/I. :) [01:21:49] More than 500 #ifexists puts the page into a category. [01:21:52] *Simetrical lures AaronSchulz away with candy [01:21:57] There's another example... [01:22:07] MZMcBride, ah, we could put it into a category. That's a viable implementation scheme. [01:22:18] And an underused one, perhaps. [01:22:30] I could do that right now except that I won't, because I'm going to sleep. [01:22:33] should be two categories [01:22:37] and one should be indexed ^_^ [01:22:38] Splarka, of course. [01:23:20] Simetrical: is it possible to categorize the pages that physically have the magic word, and not pages that transclude it? [01:23:34] so {{foo}} with __NOINDEX__ and not [[Bar]] containing {{foo}} [01:23:36] Simetrical, MediaWiki:No-index-pages could make use of regexprs like "* Wikipedia:Request_for_adminship/.*" [01:23:40] Splarka, probably possible, probably not desirable. [01:23:57] darkcode, yay, more regexes! Tim just had to disable UsernameBlacklist because its enormous regexes broke the site. :) [01:24:07] well, I mean, tracking __NOINDEX__ might be more desirable than seeing a spammy list of all pages with it [01:24:32] Splarka, except that if you see [[Template:Qulzpfstn]] on the list you have no idea that it's secretly transcluded into [[WP:NPOV]]. [01:24:52] (or, 4 categories.. [[Category:Pages manually indexed]] [[Category:Pages manually noindexed]] [[Category:Pages indexed by transclusion]] [[Category:Pages noindexed by transclusion]] ) [01:24:58] Simetrical ok LIKE statements "* Wikipedia:Request_for_adminship/%" [01:25:04] I'm looking at Extension:WikiPDF and there's an ominous message on the top of the page: "Doesn't work" [01:25:10] No other details... [01:25:25] lymeca: tell the truth, are you installing *every* extension just to try them all? [01:25:29] darkcode, how about dozens of individual pages? People who don't want their user page indexed, say? [01:25:47] This is a wiki, we don't need everything to be sysops-only. [01:26:00] Splarka: I've only asked about 2. And no I'm not, only ones which have interesting functionality. [01:26:11] __HIDDENCAT__ [01:26:16] That's the other example, I believe. [01:26:38] Yes, that's what Tim made up the idea for. [01:26:39] example of what? system-side categorization? [01:26:48] It could use some polish, like ignoring user entries into the category. [01:26:55] Splarka: Yes. [01:26:59] And maybe we could have a meta-category to include all of these, as they multiply. [01:27:13] #ifexist and __HIDDENCAT__ have categories populated based on page text. [01:27:19] well [01:27:28] not just #ifexist, any expensive parserfunctions [01:27:36] (what is the other one though.. I forget) [01:27:46] PAGESIZE? [01:27:59] The category members one? [01:28:18] probably [01:28:43] yep {{PAGESINCATEGORY}} [01:32:23] Simetrical, I'm just thinking that __NOINDEX__ would in ineffective in preventing indexing if anyone can remove it or add it, just another way to vandalise, someone decides they want to make a bunch of pages suddenly not get indexed at random, to improve there website ranking [01:36:00] maybe make it another protection level that can cascaded, where users can also add that protection level to there own user pages... [01:36:15] ugh [01:37:03] giving more edit power to users == wiki++;, giving more power to users to prevent otheres from editing == wiki--; [01:37:23] 03shinjiman * r38210 10/ (10 files in 4 dirs): [01:37:23] Update the Chinese conversion tables [01:37:23] Last minute changes for the release branch 1.13 [01:39:24] its not really editing related though, its just indexing related [01:40:13] still, it does reduce editing incentive [01:40:49] Simetrical: what do you think about general magic word tracker categories for all "senisitve" magic words, like __HIDDENCAT__ __INDEX__ __NOINDEX__ __NEWSECTIONLINK__ __NOGALLERY__ {{int:xyz}} {{DISPLAYTITLE:xyz}} etc [01:41:20] and (hypothetically, though the code might be dirrrrrrty) pages that have it literally could be defaultsorted as *, those transcluding could have it alphabetically [01:41:56] and then instead of custom categorizing hacks for each new one, there could be a simple system in place [01:44:16] (per silly idea, as if each __NOINDEX__ page had [[Category:Noindexed pages|*{{FULLPAGENAME}}]]) [01:44:19] imo, that looks doable, though not sure of the use [01:44:51] man, I should eat something [01:46:55] Simetrical: hey, wait a minute. I don't really like candy! [01:47:18] *AaronSchulz took 24 minutes to notice that [01:55:48] 03shinjiman * r38211 10/ (2 files in 2 dirs): [01:55:48] Localisation updates Chinese [01:55:48] Last minute changes for the release branch 1.13 [02:38:28] 03(mod) Edit form eats heading CRs on save/preview - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=12130 (10jayvdb) [02:39:47] 03(mod) New revisions occasionally created with wrong text ( but correct rev_len) - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14933 +comment (10herd) [02:56:05] I've got 1.13.0rc1 on the latest stable Apache and PHP versions, and I've installed the FCKeditor extension and it works. But there are pretty severe performance issues and I don't know how to troubleshoot them. [02:56:31] In Firefox 3, it takes at least 10 seconds for the editor to load up in the page [03:04:09] <^demon> Any other extensions you're running in addition to that? [03:05:36] 03demon * r38212 10/trunk/phase3/includes/ (ImageQueryPage.php OutputPage.php): Remove unused $wgServer global. [03:08:06] 03demon * r38213 10/trunk/phase3/includes/ImageQueryPage.php: Bah! Partial revert of r38212. Unrelated file snuck into my commit window. [03:16:19] 03(mod) Edit form eats heading CRs on save/preview - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=12130 (10jayvdb) [03:18:39] <^demon> Simetrical: Ping [03:19:06] He's definitely asleep. :) [03:20:03] <^demon> Gotcha ;-) [03:20:46] <^demon> He was asking what the ssl-login-work branch was earlier. I just happened across the answer a few minutes ago :-P [03:21:59] <^demon> Brion was doing some testing with $wgServerName for bug 8866 like a year ago :-P [03:23:51] 03demon * r38214 10/trunk/phase3/ (RELEASE-NOTES includes/DefaultSettings.php): (bug 14977) Installations on servers using IPv6 addresses for $wgServer were largely unusable because $wgServer didn't wrap the IP in brackets [03:25:36] 03(FIXED) $wgServer lacks brackets in IPv6 URLs - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14977 +comment (10innocentkiller) [03:25:49] 03(FIXED) An extension to create list of most recent articles in 1-3 categories. - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=1411 +comment (10innocentkiller) [03:26:11] Heh. [03:26:22] Fixing the already-fixed bugs, eh ^demon? ;) [03:27:11] <^demon> Yeah, I got lazy. [03:27:30] also called "scraping squished bugs off other people's boots for your collection" [03:27:58] <^demon> Splarka: I take no credit, I just close it so the number of total bugs goes down :-P [03:31:39] hello. what is the "correct" way to link to the "sign in / create account" page? [03:32:35] [[Special:Userlogin]] [03:32:46] thank you [03:33:40] *qsheets snickers [03:33:58] well [03:34:20] there is always a better way ^_^ [03:34:20] is there a way to use the one of the parsing extensions to determine if a user is already logged in? i.e., if they're logged in, i want them to view the page normally; if not, i want to add a line saying "please register or login" [03:34:34] {{#special:UserLogin}} [03:34:49] will translate to the special page default name, depending on content language [03:35:24] DanielTahar: no, that would break caching badly (see wontfixed bugs for USERNAME or CURRENTUSER magic words) [03:35:29] <^demon> I am convinced that Special:Allmessages is more useful to developers than end-users :-) [03:35:41] but you can do that with an anon sitenotice [03:35:59] <^demon> Splarka: It could be done with Javascript and the wgUser variable. [03:36:06] unreliable [03:36:21] DanielTahar: do you want to actually restrict their ability to view the page, DanielTahar? [03:36:27] mm, redundancy, oops [03:36:41] or he can make an extension w/ the wgUser variable [03:37:02] *Splarka nods [03:37:06] perhaps a modification of sitenotice [03:37:29] erm, can just use anon notice if that is the only change [03:37:38] there's [[MediaWiki:Anonnotice]] [03:37:58] *Splarka only said that twice, heh [03:38:05] lol [03:38:24] well, not really [03:38:27] it's no biggie [03:38:42] note that sitenotices are cached oddly, things like {{#SWITCH:{{NAMESPACE}}}} or {{PAGENAME}} won't work as you predict [03:38:45] i'll just use a line saying "don't forget to login or register" [03:38:59] DanielTahar: add it to [[MediaWiki:Anonnotice]] [03:39:22] you could even make it namespace specific (not with #switch) [03:39:41]
Register, losers!
in MediaWiki:Anonnotice [03:39:58] #remindanonstoregister {display:none;} [03:40:06] body.ns-0 #remindanonstoregister {display:block;} [03:40:12] would have it only show in article namespace [03:42:49] thanks! [03:55:33] 03(mod) Image hotlinking from squatted domain - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14829 +comment (10lilewyn) [04:04:58] ^demon: Yes I am running other extensions alonside FCKeditor. [04:05:47] LDAP Authentication Plugin and Cite [04:13:44] 03demon * r38215 10/trunk/phase3/ (RELEASE-NOTES includes/specials/SpecialSearch.php): The documented setting on MW.org (and in DefaultSettings) is $wgSearchForwardUrl, so tweaked Special:Search to use the newer (and better) titled setting. Should make it actually work as advertised now. [04:15:34] 03demon * r38216 10/branches/REL1_13/phase3/ (RELEASE-NOTES includes/specials/SpecialSearch.php): Backport fix from r38215 to branch. [04:15:39] <^demon> lymeca: I know there's been some issues with FCKEditor and recent versions? I'm not 100% on that :( [04:18:51] hmm [04:22:50] ^demon: What's the best way to delete a user from the wiki? [04:23:09] user merge and delete extension [04:23:41] if you absolutely *have* to delete someone, otherwise I'd just leave 'em there [04:24:07] <^demon> What he said ^ :-) [04:25:00] best way is not to, ^_^ [04:33:27] *Splarka wonders if the API should have an indicator in props=categoryinfo for __NOGALLERY__ [04:35:02] http://commons.wikimedia.org/w/api.php?action=query&prop=categoryinfo|revisions&rvprop=content&titles=Category:Maps%20in%20Spanish|Category:The_Ashes,_1882 <-- eg, this works, but only if it isn't transcluded [04:38:30] <^demon> Just for my own curious interest, why does protection code require a dummy revision be put into the database? [04:39:44] to show on watchlist, history, contribs? [04:40:11] (although it broke in watchlists for the longest time) [04:41:02] <^demon> I figured if it made a log entry, why bother with a revision? Other logged actions don't do that. [04:41:24] move does [04:41:32] and log entries don't show on history/contribs [04:41:46] It's always made me sad that move uses a Special: page. [04:42:09] <^demon> MZMcBride: You'd prefer action=move? [04:42:14] Yes. [04:42:57] *^demon shrugs. [04:46:54] it always made me sad that log actions against users and user pages were treated the same [04:47:17] you block [[User:Splarka]] and protect [[User:Splarka]], you delete [[User:Splarka]] and promote [[User:Splarka]] in userrights, etc [04:47:57] (for the purposes of Special:Log that is) [04:49:26] Yes, that's horribly confusing as well. [04:49:40] Especially with the text boxes Title: and User:. [04:49:42] maybe for new users [04:49:46] *Splarka wonders how to tell if a category has __NOGALLERY__ [04:49:56] AaronSchulz: Not just new users. [04:49:58] yeah, having to say User: is prolly the worse part [04:50:18] I can tell per-page via getElementsByClassName mw-category-media, but.. meh [04:50:47] *AaronSchulz plays with DPL [04:50:54] Would a magicwordlinks table be a terrible idea? [04:51:02] what for? [04:51:11] To track magic words. [04:51:22] it is doable without too much drain [04:51:31] but would it be worth it? [04:51:47] not sure since I don't know what one would do with the lists [04:51:54] I think the previous discussion about categorizing more magic word usage, like __HIDDENCATS__, is better [04:52:22] eg __INDEX__ __NOINDEX__ __NOGALLERY__, maybe a few others like displaytitle [04:52:35] yeah, that's what I thinking of [04:52:38] <^demon> It'd be similar to the IW table, I guess. The question is: do you want to query the magicwords table on every pageview so you can then run them through the parser? I guess it could be cached _very_ heavily... [04:53:03] ^demon: not any worse the pagelinks lag [04:53:08] <^demon> True. [04:53:18] like if a template is conditionally included and has cats [04:53:32] the cat data would get lagged if the condition changed [04:53:44] well, for things like time and stuff [04:54:12] heh [04:55:07] what would {{#ifexpr: {{CURRENTDAY}} mod 3 > 1 | __HIDDENCAT__ }} do I wonder [04:56:42] the table would be stuck on the state it was in during the last edit [04:57:11] or null edit, or purge, or edit to any templates transcluded on it? [04:57:35] I could null edits as 'edits' ;) [04:57:40] fair [04:57:41] not purge I believe [04:57:50] nor edits to templates [04:58:05] only very select magic words would need to be categorized or tabled, like those used once per page if at all, and rarely transcluded [04:58:06] edits to templates used call HTMLCacheUpdate which does purge on all pages used [04:58:13] but purge alone doesn't do linksupdate [04:58:44] Splarka: I think the parser output expire length is different for some pages with time magic words [04:58:52] still pretty hacky though [04:59:06] <^demon> My database is so messy now, I should probably redo my local install. Various funny bits here and there due to introducing/fixing bugs. [04:59:12] mm, hacky [04:59:37] ^demon: yeah, mine got pretty messed up at times [04:59:47] man, I love my rice cakes [05:00:06] like a bunch of page rows that pointed to revs that were not there [05:00:27] which leads to some *very* interesting behavoir [05:02:06] <^demon> AaronSchulz: Question about FlaggedRevs. When viewing search results, do the blurbs respect the sighted versions or not? [05:02:34] if a page has a sighted version, and it shows by default, only that version is indexed [05:02:47] well, assuming robots stuff is followed :) [05:03:33] <^demon> I meant the MW internal search :-) [05:04:05] lol [05:04:11] no, it is still the same [05:04:25] <^demon> Gotcha. [05:05:16] Splarka: well I suppose a null edit isn't an 'edit' in any ordinary sense of the word, since the page is the same and there is no history [05:05:50] mmm, I remember one fun bug [05:05:51] *AaronSchulz likes to indulge in jargon howeve [05:05:58] 03(NEW) Checkuser should display expired blocks as well as active ones - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14979 15enhancement; normal; MediaWiki extensions: CheckUser; (mapellegrini) [05:06:00] some of the revision-specific timestamp magicwords... [05:06:02] *however [05:06:06] they'd change to the current time in null edit [05:06:14] but go back to last actual edit time with action=purge [05:06:15] People make trivial edits and use the summary "null edit" sometimes. That makes me sad. [05:06:18] *AaronSchulz lols at bug 14979 [05:06:31] heh, I said I was going to work on recursive checkuser months ago [05:07:26] 03(mod) Blocking and tagging from checkuser interface - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=12808 normal->highest; +comment (10mapellegrini) [05:07:31] <^demon> And I keep saying I'm rewriting blocking. [05:07:52] 03(mod) Make DPL extension aware of FlaggedRevs for use on Wikinews - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14943 (10JSchulz_4587) [05:10:23] How are the passwords hashed in the DB? [05:10:25] Is it MD5? [05:10:33] It used to be. [05:10:52] It was md5(userid - md5(password)) [05:10:56] But I think it changed recently. [05:11:13] <^demon> AaronSchulz: Just tried to install FlaggedRevs on my fresh install. Update.php crashed. [05:11:24] lol [05:11:50] <^demon> plaggedpages table :-) [05:11:54] <^demon> *flagged [05:12:21] MZMcBride: Ah, well I'm using 1.13.0rc1 now, but I only use LDAP now so apparently passwords aren't stored in the DB which makes sense. But when I wasn't using LDAP I had 1.11.0 [05:12:34] So it depends on the MediaWiki version? Did 1.11.0 use MD5 hashing? [05:12:43] Yes. [05:14:24] ^demon: it shouldn't have to run anything there if it is fresh [05:14:49] <^demon> I was running update.php to make flaggedrevs tables appear ;-) [05:16:11] <^demon> http://pastey.net/92737 - there's the error, be right back. [05:17:10] ahh, did you run the flaggedrevs.sql query first? [05:19:48] anyone know how to access a page by it's ID number? [05:20:15] Access how? [05:20:52] normally - like instead of ?title=pagename, ?id=1003 [05:21:11] Try ?old=### [05:21:15] thanks [05:21:16] Oop. [05:21:20] ?oldid=### [05:21:23] ok [05:21:48] you might also need a null title [05:21:55] &title=-&oldid=123 [05:22:13] sometimes you get the main page without that [05:22:33] <^demon> AaronSchulz: I was under the impression that update.php added the tables :-P [05:22:41] nope [05:22:41] didn't seem to need a null title [05:22:48] <^demon> That would be why :-P [05:22:51] thanks, works well [05:24:51] Anyone able to help me figure out what I'm doing wrong with the DISPLAYTITLE magic word? [05:25:52] if you're using >1.12 of MW it will only change capitalization [05:26:06] oh, I see [05:27:38] OverlordQ: sort of wrong and sort of wrong [05:27:45] I want to write an extension tag for my mediawiki instance, but within the extension I want to work out the name of the wiki that called it. Is there a global variable that can tell me that? [05:27:53] more specifically, it only changes the title in ways that make it still linkable by copypaste [05:27:58] well, acutally, I'm using 1.11.0 [05:28:27] eg, [[Foo Bar]] can be [[foo_Bar]] [[foo Bar]] [[Foo_Bar]] but not [[foo BAR]] [05:28:28] lol :) [05:28:58] and it might allow and or some other shit, or someone was requesting that [05:29:05] i know there was words like normalization in there somewhere but too lazy to remember correctly [05:29:12] heh [05:29:13] what about in version 1.11.0? [05:29:27] well, if you use it for that, it'll break when you upgrade ^_^ [05:29:53] how drastically do you need to change the titles? [05:30:23] So, was the original function completely trashed, or is there a new way to call it? [05:30:33] completely different :P [05:31:05] 03aaron * r38217 10/trunk/extensions/FlaggedRevs/ (FlaggedArticle.php FlaggedRevs.class.php): Improve and use getTagMsg() [05:31:23] <^demon> AaronSchulz: Works great now :-P [05:31:56] DuTempete: well, the original function was sorta silly... [05:31:59] be sure to try $wgGroupPermissions['*']['feedback'] = true; [05:32:36] <^demon> AaronSchulz: What's that do? [05:32:39] DuTempete: a lot of editors, a large percentage anyway, link to a page by copying the h1 class=firstHeading [05:32:50] ^demon: [05:32:59] if you mess with that, like by adding extra spacing, line returns, or changing it drastically, people complain [05:33:04] *AaronSchulz injected some test data [05:33:12] are you wanting to change the

or the or both? [05:33:27] <DuTempete> basically, I want a page ".../nameyyyymm" to display as "month of year, for name" [05:33:28] <^demon> Ooh, pretty :-) [05:33:37] <DuTempete> Splarka, my site is not public [05:34:06] <AaronSchulz> ^demon: the girl or the graph? ;) [05:34:12] <jdownie> ???I want to write an extension tag for my mediawiki instance, but within the extension I want to work out the name of the wiki that called it. Is there a global variable that can tell me that? [05:34:16] <Splarka> DuTempete: looks like it was changed to the normalizing method in 1.10 [05:34:19] <Splarka> http://www.mediawiki.org/wiki/Manual:%24wgAllowDisplayTitle [05:34:20] <^demon> AaronSchulz: Graph. [05:34:28] <Splarka> http://svn.wikimedia.org/viewvc/mediawiki?view=rev&revision=23393 [05:34:39] <Splarka> so if you have revision higher than 23393, you can't even use it on yours as you want [05:34:42] <DuTempete> There will be 6 regular editors to begin, with a possibility of 8 in the future :P [05:34:52] <Splarka> you'll have to use javascript [05:35:07] <Splarka> or an extension with another parser tag hook or parser function [05:35:20] <Splarka> something like {{#realtitle:}} [05:35:21] <DuTempete> gotcha [05:35:37] <Splarka> that actually shouldn't be too hard, read up on how to make an extension ^_^ [05:35:54] <DuTempete> I don't have the time [05:36:32] <DuTempete> I'm designing a family wiki, with the main function being as a calendar, in my spare time atm :P [05:36:39] <Splarka> could you live with redirects? [05:36:43] <DuTempete> nope [05:36:49] <DuTempete> doesn't solve my problem [05:36:51] <Splarka> ".../nameyyyymm" would still be the URL [05:36:56] <Splarka> you could hide 'redirected from' [05:37:29] <Splarka> ahh well, go with a hacky JS solution for now, switch to extension later if anyone can write it for ya [05:38:13] <DuTempete> I want my mother to know what page she is editing without having to think to hard about what the name means, whilst still being able to alphabetise the pages automatically within the category :P [05:38:26] <Splarka> eg, if you make {{realtitle|foobar}} with content <div id="realtitle-override">{{{1}}}</div> ... you can change the content of that template later to {{#realtitle:{{{1}}}}} when you have an extension [05:38:56] <Splarka> and until then, use an addOnloadHook(function() { ... }) that checks for document.getElementById('realtitle-override') [05:39:30] <Splarka> and sets document.title and getElementsByClassName(document,'h1','firstHeading')[0] [05:40:14] <Splarka> like 5 lines of code [05:40:24] <DuTempete> *nod* Thanks much. [05:40:31] <Splarka> rar [05:40:41] <DuTempete> I have a buddy who is far better with JS than I am. I'll make him do it fo rme :P [05:40:49] <Lady_Aleena> zip :) [05:40:59] <Splarka> heh [05:41:00] <DuTempete> That stuff gets a bit too advanced for me! [05:41:08] <lymeca> Does anyone know where to find the e-mail address of Ryan Lane? [05:41:12] <Splarka> well, be sure to tell him to use addOnloadHook (built in), easy [05:41:57] <Lady_Aleena> Splarka, sorry, couldn't help myself, the very first thing I saw when comig back in was "rar" and the first thing that popped into my head was [05:41:59] <DuTempete> yeah, he alreday uses that. He's written quite a fair chunck of JS for a particular mediawiki site [05:42:01] <Lady_Aleena> "zip". [05:42:08] <Splarka> silly lady [05:42:15] <Splarka> Du: sweet, then he can do it no prob [05:42:36] <DuTempete> yeah, probably have it written in about 10 minutes. [05:42:38] <Lady_Aleena> lymeca, have you gone to his user page and tried his email link? [05:43:11] <lymeca> Lady_Aleena: http://www.mediawiki.org/wiki/User:Ryan_lane I don't see an e-mail link? [05:43:19] <Lady_Aleena> There is a nice little link in the toolbox "Email this user" [05:43:33] <Lady_Aleena> "E-mail this user" to be exact. [05:43:59] <Lady_Aleena> It is right beneath the link to "Logs" [05:44:25] <lymeca> Lady_Aleena: Do I have to be logged in or something? [05:44:27] <lymeca> I don't see it. [05:44:33] <Splarka> heh yes [05:44:33] <Lady_Aleena> http://www.mediawiki.org/wiki/Special:EmailUser/Ryan_lane [05:44:44] <Splarka> and you have to have your email set and confirmed too, probably [05:45:09] <lymeca> Once logged in can you tell what the address is? [05:45:21] <lymeca> Or do is it hidden and you have to send through the wiki? [05:45:23] <Lady_Aleena> lymeca, nope, but you can send the e-mail. [05:45:33] <lymeca> Honestly I'd rather use Thunderbird to communicate with him. [05:45:36] <lymeca> I see. [05:46:33] <Lady_Aleena> lymeca, the first message should go through the wiki, then if he cares to respond, then he can directly to your e-mail address. [05:46:47] <lymeca> Okay, makes sense. [05:46:54] <Lady_Aleena> Then you can use your e-mail client all you wish. [05:47:13] *Lady_Aleena uses Outlook. [05:50:11] <Lady_Aleena> Okay, I am trying to build a template, but something has gone really really wrong. Any template gods here want to help me debug it? The template is http://en.wikipedia.org/wiki/User:Lady_Aleena/Test_banner and to see what is wrong look at http://en.wikipedia.org/wiki/User_talk:Lady_Aleena/Test_banner pretty please? [05:51:51] <Lady_Aleena> And Splarka, take note, someone said that I made sense. :) WooHoo! [05:55:29] <mkv-> Morning. [05:55:34] *Splarka alerts Ripley's for LA [05:55:54] <Lady_Aleena> Splarka, not good enough for Ripley's quite yet. [05:56:33] <lymeca> Outlook? Ewww... [05:57:01] <Lady_Aleena> lymeca, do you know how much of a culture shock I went through when I first switched from IE to Firefox? [05:57:40] <lymeca> Lady_Aleena: Howdoes that have anything to do with culture shock? Did you suddenly become involved in the Free Software community with that transition? [05:58:02] *Lady_Aleena still can't find where she screwed up in that template, and this her upteenth look throught. [05:58:15] <lymeca> I mean there's an underlying cultural difference with respect to licensing ideologies, but most of it is just technological superiority. [05:58:32] <Lady_Aleena> lymeca, IE and Firefox are so different that it took time for me to adjust. [05:58:56] <lymeca> Just pointing out that it's not much to do with culture. [05:59:08] <Lady_Aleena> lymeca, I was using that metaphorically. [05:59:19] <lymeca> Also mayne people who used the web in the mid-90's experienced a switch from Netscape to IE. [05:59:50] <lymeca> But ultimately, a browser UI is a browser UI. I find them all to be very similarly usable. [06:00:28] <lymeca> Outlook and Thunderbird are somewhat more different from each other than IE and Firefox though. [06:00:34] <Lady_Aleena> lymeca, do you know how much angst I went through when I first saw my web site rendered on Firefox? I was in shock. [06:00:39] <lymeca> Lurning curve is worth it for freedom. [06:00:48] <mkv-> lymeca: What about Evolution then? :) [06:01:14] <lymeca> Lady_Aleena: Do you know how much angst basically any web developer goes though trying to support older IE versions and their crippling of web standards? [06:01:31] <Splarka> my first email was in ~1996, it told me to get Netscape Navigator 2.0 and ditch Explorer [06:01:47] <lymeca> mkv-: Not sure, haven't used it much. [06:01:58] <Splarka> "don't fool with Explorer unless you hate life." [06:02:00] <Lady_Aleena> lymeca, all I know is that when I saw my web site for the first time on Firefox, I cried for about half an hour wondering how in the world I was going to fix it. [06:02:07] <mkv-> lymeca: Okay. It's quite similar to Outlook. A bit buggy still but it works. [06:02:19] <lymeca> But yeah there are many options. Thunderbird, Evolution, KMail, (pine ;-) [06:02:23] <mkv-> lymeca: Quite good even with a SBS. [06:02:31] <mkv-> lymeca: pine *shrugs* ;) [06:02:32] <lymeca> A lot of upgrade paths from Outlook [06:02:36] <mkv-> Long time no see. [06:02:38] <lymeca> hehe [06:03:22] <lymeca> Lady_Aleena: I've a suggestion: Code your site to W3C standards and through a badge recommending your web users to download a standards-compliant browser such as Firefox. That's what I do on my sites. [06:03:25] <Lady_Aleena> I still wish that I could use Perl on my site to make it even smaller, but I have yet to find someone to help me do it. (Basically teach me as I go and maybe write some of it.) [06:03:33] <lymeca> *throw [06:03:44] <mkv-> Anyway, I should keep configuring my mediawiki.. Quite much to think about.. (and nothing to think with :)) [06:03:54] <lymeca> mkv-: Yeah I'm doing that too. [06:04:05] <lymeca> I'm installing/configuring extensions [06:04:11] <Lady_Aleena> lymeca, if you go to my user page on en, you can get to my website, and see that I have these two little images showing my site to be W3C compliant. [06:04:19] <mkv-> lymeca: Oh, I'm only in the beginning of the configuration.. *sigh* [06:04:43] <lymeca> mkv-: Heh, good luck. Be sure to make backups!! I had to start from scratch this past weekend =[ [06:04:59] <mkv-> lymeca: Ouch. [06:05:08] <Nikerabbit> ^demon: -m is bad [06:05:13] <mkv-> I'm going to write own extensions also. [06:05:50] <lymeca> Lady_Aleena: That's good, especially if they pass. I generally honestly don't care if IE breaks my site or not. I'm not going out of my way to break my site and support the application which is singularly responsible for holding back web standards adoption. [06:05:56] <lymeca> SVG what wwhhaatt? [06:05:59] <Lady_Aleena> Anyone know any good forums for getting help with Perl which doesn't hit newbies over the head with the Camel book? [06:06:02] <mkv-> lymeca: Aint there a admin-interface anywhere to be found? [06:06:08] <mkv-> Lady_Aleena: perl.org [06:06:11] <OverlordQ> perlmonks [06:06:29] <lymeca> mkv-: Admin interface for what? I've installed a few extensions which make sysop users much more powerful. [06:06:33] <Lady_Aleena> OverlordQ, Perlmonks hits me over the head with the Camel book which is too damned expensicve. [06:06:37] <Lady_Aleena> expensive. [06:06:57] <mkv-> lymeca: Ah, those would be great. Did you download the extensions from mediawiki or anywhere else? [06:07:15] <lymeca> Nope, just mediawiki [06:07:20] <lymeca> There's plenty to choose from =] [06:07:29] <lymeca> Unfortunately my FCKeditor has wretched performance. [06:07:31] <mkv-> lymeca: Great. :) Thanks. I'll look into it right away. [06:07:44] <Lady_Aleena> mkv-, do they have a forum there where one can say this is what I want to do and there are people who are willing to show me where I messed up for free? [06:07:48] <lymeca> OverlordQ: Do you have experience with either FCKeditor or OggHandler? [06:08:11] <mkv-> Lady_Aleena: Hmm, dont know about the forum. But there is lots of help there. [06:08:26] <mkv-> Lady_Aleena: Better if you find the problem yourself. :) [06:08:59] <Lady_Aleena> mkv-, I need a site which for Perl which is equivalent to the W3C site where EVERYTHING is spelled out nice and simple for newbies and experts alike. [06:10:21] <Lady_Aleena> I have yet to find a page anywhere that tells me how to write a complex random generator or build a table from a database, both to be output onto a web page. [06:10:46] <^demon> Nikerabbit: Huh? [06:11:10] <mkv-> Lady_Aleena: Okay. I do not know then sorry. [06:12:01] <Lady_Aleena> I tried on Perlmonks, I really did, but there was no one to hold my hand through the process of learning like on another forum where I got help and then some for writing my initial website. [06:12:08] <Nikerabbit> ^demon: tends to sneak unrelated files [06:12:08] <Splarka> http://www.cracked.com/article_16521_5-tiny-mistakes-that-lead-huge-catastrophes.html wonder when Wikimedia will make the list [06:17:01] <Werdna> Lady_Aleena: Generally speaking, you should teach yourself, not expect others to do it for you. [06:17:32] <Lady_Aleena> Werdna, I was trying, but there is not enough information on the internet, that I could find to learn Perl inside and out. [06:17:47] <mkv-> lymeca: What extensions do you recommend for sysop handling? [06:18:12] <lymeca> mkv-: It really depends on what you want, and I have by no means read every entry in the Extensions Matrix... [06:18:24] <lymeca> but... [06:19:48] <Werdna> Lady_Aleena: generally speaking, find a project, and motor along and learn as you go. [06:19:54] <Werdna> That's how I learned perl, anyway. [06:20:19] <mkv-> Werdna: Word. [06:20:28] <mkv-> Werdna: Learn by living. :) [06:20:36] <Lady_Aleena> Werdna, my first project, which has stalled, was building a vast random generator. [06:20:57] <lymeca> mkv-: So far I have User Merge and Delete, User Snoop, WhoIsWatchingTabbed, and LDAP Authentication Plugin for the backend user auth to give me more control over accounts. [06:22:22] <mkv-> lymeca: Ok. I do not use LDAP (yet anyway, maybe I should?). [06:22:41] <lymeca> mkv-: I mostly chose it because my LAN already had an LDAP server running. [06:22:57] <lymeca> But I do find it gives me much more flexibility, as long as you have write access to the LDAP server =P [06:23:14] <mkv-> lymeca: Okay. This is only one server running. So no LAN behind or anything. [06:23:24] <mkv-> lymeca: This is for official purposes. [06:23:32] <lymeca> oh, wel technically you can run the LDAP server on localhost [06:23:37] <mkv-> Hopefully ~10.000+ more registered users. [06:23:59] <lymeca> Humm, there's nothing saying that a wiki server on a LAN can't be official or have lots of users... [06:24:11] <mkv-> lymeca: Yes, I know. But I do not know if I win anything by creating a LDAP-db instead of using the builtin MySQL for user-information. [06:24:13] <lymeca> In fact I'd bet most wikis are on LANs. [06:24:46] <lymeca> mkv-: Right, I don't think I would've started an LDAP server just for the wiki. I have other services which use LDAP though so it's convenient in that respect. [06:25:46] <mkv-> lymeca: Yea, just thought if I would gain performance or something. [06:25:57] <mkv-> But then I'll just put LDAP to the side for the moment. [06:28:36] <mkv-> Uh I need to understand the user rights in mediawiki first. [06:29:59] <^demon> Lady_Aleena: Looking for a Perl project to get your hands on to learn some stuff? [06:30:28] <Lady_Aleena> ^demon, not right now, at the moment I have a template that is driving me nuts. [06:31:26] <^demon> Lady_Aleena: If you ever have the time and want some code to play with, I know a Perl-based message board run by a few friends of mine (I help out from time to time) who would love a new dev to lend a hand :-) [06:31:35] <Lady_Aleena> The blasted thing is not closing the table after a few parameters are set. [06:32:03] <Lady_Aleena> ^demon, I would have to relearn that few basics that I have now forgotten. [06:33:00] <Lady_Aleena> The only thing I remember is that a scalar starts with a my $, and there are arrays and hashes and things. [06:41:16] <^demon> What would be the easiest configuration variable(s) to find the path to the MW api? [06:41:41] <qsheets> ^demon: ?? [06:42:26] <Splarka> the scriptpath? same for thumb.php and index.php ? [06:42:37] <^demon> That would be it :-P [06:44:24] <wikibugs> 03(mod) escapeshellcmd does not work properly with php security update - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14944 (10daniel) [06:57:36] <justin2net> hi, could anybody help me setting up email notificaitons? [07:11:58] <sentinel100101> simply put what are breadcrumbs extension how does it help? [07:19:26] <CIA-55> 03purodha * r38218 10/trunk/phase3/languages/messages/MessagesEn.php: Superfluous unneccessary html markup removed from 3 messages. [07:23:35] <CIA-55> 03vasilievvv * r38219 10/trunk/phase3/includes/api/ApiMain.php: * (bug 14968) Specifying blank action parameter triggers API error now [07:24:29] <wikibugs> 03(FIXED) Specifying blank action parameter triggers WMF Error instead of API Error - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14968 +comment (10vasilvv) [07:25:03] <sentinel100101> what does the bread crumb extension do? [07:30:35] <CIA-55> 03huji * r38220 10/trunk/phase3/languages/messages/MessagesFa.php: Localisation updates: Adding/updating Persian translations [07:30:42] <sentinel100101> !breadcrumb [07:30:42] --mwbot-- I don't know anything about "breadcrumb". [07:31:30] <guyvdb> what's the point of all the "@ingroup" annotations? [07:45:37] <^demon> guyvdb: They're for the PHPDoc stuff. [07:46:39] <CIA-55> 03demon * r38221 10/trunk/phase3/ (5 files in 4 dirs): Add new parser function {{apiurl}}. Also, add new global $wgApiScript because manually constructing the api script's path seemed like a bad idea. [08:02:50] <ajki> hello, is someone here? i ve got problems with Mediawiki on Mac OS X Server (installed "into" Xampp) [08:03:35] <ajki> The problem is an crashed searchindex, and i cant repair it. The table is crashing all the time again, after i am editing one page in the wiki. [08:08:35] <ajki> is someone here? Yes i read the FAQ.. [08:08:47] <OverlordQ> sounds like a problem unrelated to mediawiki, might . . . [08:08:48] <OverlordQ> or not [08:09:17] <lymeca> OverlordQ: Are you familiar with the error: [08:09:30] <lymeca> Warning: xcache_get() [function.xcache-get]: xcache.var_size is either 0 or too small to enable var data caching in includes/BagOStuff.php on line 607 [08:09:49] <OverlordQ> need to set an actual side in the xcache config [08:09:51] <OverlordQ> *size [08:10:24] <lymeca> I altered the xcache.var_size to say 64M in /etc/php5/conf.d/xcache.ini [08:10:28] <lymeca> Then I restarted Apache [08:10:32] <lymeca> But I'm still getting the error. [08:11:09] <OverlordQ> built from source or the package? [08:11:23] <lymeca> xcache.var_size = 64M [08:11:26] <lymeca> package [08:11:49] <lymeca> I will note that I installed the wiki when it was at 0M [08:12:03] <lymeca> It's a new wiki, so reinstallation wouldn't be bad [08:12:57] <OverlordQ> shoul'dn't matter [08:13:12] <lymeca> Okay, well a vlue of 64M worked on another wiki of mine [08:13:19] <lymeca> Any idea why the errors persist? [08:14:00] <OverlordQ> is there a symlink in /etc/php5/apache2 named conf.d that points to ../conf.d [08:14:35] <lymeca> Yes [08:15:54] <OverlordQ> I dunno, check xcache.size [08:16:45] <lymeca> 16M [08:16:52] <lymeca> I'm going to reinstall [08:17:15] <OverlordQ> _count isn't 0? [08:17:49] <lymeca> xcache.count = 1 [08:17:55] <lymeca> xcache.slots = 8K [08:18:05] <lymeca> xcache.var_count = 1 [08:18:05] <lymeca> xcache.var_slots = 8K [08:18:06] <OverlordQ> what about xcache.var_count [08:18:18] <OverlordQ> strange, double sure you restarted apache? [08:18:23] <lymeca> triple sure [08:18:44] <OverlordQ> dunno then, wfm [08:21:54] <lymeca> OverlordQ: reinsallation worked [08:22:05] <lymeca> Ran the config.php script again from a fresh 1.13.0rc1 tarball [08:24:08] <OverlordQ> *shrug* odd [08:33:20] <wikibugs> 03(mod) Set $wgBlockAllowsUTEdit for Greek Wikipedia - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=10304 15enhancement->04CRIT; normal->high; +comment (10geraki) [09:17:37] <CIA-55> 03siebrand * r38222 10/trunk/phase3/languages/messages/ (39 files): Localisation updates for extension messages from Betawiki (2008-07-30 11:03 CEST) [09:25:55] <Tony1> hello [09:27:02] <Tony1> a quick question please: is this the right place to ask question related to mediawiki installation? [09:28:09] <CIA-55> 03siebrand * r38223 10/trunk/extensions/ (65 files in 59 dirs): Localisation updates for extension messages from Betawiki (2008-07-30 11:03 CEST) [09:40:21] <Theblase> Hi, [09:41:13] <Theblase> I have a fresh MediaWiki installation (1.12.0). I'm a beginner and want to switch language directly on the main page like the attached image. [09:41:29] <Theblase> http://wiki.typo3.org/index.php/Main_Page [09:41:47] <Theblase> I install the Extension:LanguageSelector, but this is not the same that used in the http://wiki.typo3.org/index.php/Main_Page, in the top right corner. [09:41:57] <Theblase> Can anyone help me? [09:44:00] <Jhs> Theblase, maybe [09:44:23] <Theblase> Many thanks! [09:44:58] <Theblase> I waiting for your instructions. [09:47:08] <Werdna> domas: if I have a pointer and a size_t of what it points to, how do I free the memory [09:47:19] <Werdna> checking 250,000 moves currently uses 70% of my memory :) [09:47:59] <Jhs> Theblase, the thing is that typo3 doesn't use an extension at all, they use a (slightly complicated) template [09:48:06] <VasilievVV> Werdna: free()? [09:48:56] <Werdna> that takes just a pointer [09:49:00] <domas> Werdna: thats usually enough [09:49:01] <Werdna> and has undefined behaviour for just a void* [09:49:12] <domas> Werdna: free() works on same pointer that was given by malloc() [09:49:21] <Theblase> Ok, I understand it. [09:49:23] <domas> if you're using objects, of course you do 'delete' [09:50:07] <Werdna> domas: in the object's destructor (does that get called when the object goes out of scope?), I want to free a void*, which points to either string, long int or double. I have a size_t member so that I know how big it is. [09:50:09] <Theblase> Have you got any idea for this language switsching? [09:50:28] <VasilievVV> Werdna: how have you allocated it? [09:50:29] <domas> how did you get that void* ? [09:50:43] <domas> and no, object destructors are probably not called :) [09:50:48] <domas> unless you use some GC [09:51:18] <Werdna> domas: I have a class which can store ints, strings or floats. [09:51:26] <Werdna> so I have a void* pointer to the data, a type member and a size member [09:51:37] <Theblase> What is the necessary extension to language switching? [09:52:55] <domas> Werdna: well, delete it then [09:52:58] <domas> once you're done :) [09:53:30] <Werdna> do I need to cast to the appropriate pointer type before I do that? [09:54:18] <Werdna> aftypes.cpp: In destructor ??????AFPData::~AFPData()??????: [09:54:18] <Werdna> aftypes.cpp:136: warning: deleting ??????void*?????? is undefined [09:58:55] <Werdna> *** glibc detected *** /home/andrew/mediawiki-extensions/AbuseFilter/parser_native/af_parser: double free or corruption (fasttop): 0x080781d8 *** [09:58:59] *Werdna stabs people [10:00:46] <VasilievVV> Werdna: idea is probably that the memory used by passed data should be released by code that uses AFPData, not by AFPData itself [10:04:43] <Werdna> Freeing 0x80781d8 - type 1 [10:04:43] <Werdna> Freeing 0x80781d8 - type 1 [10:05:02] <Werdna> It's an integer [10:08:26] <CIA-55> 03raymond * r38224 10/trunk/extensions/ (5 files in 5 dirs): Add a few punctuation marks (.) for consistency with other header/summary messages. [10:12:41] <Extend> hey guys i wanna ask about something in mediawiki [10:12:50] <Werdna> impressive! [10:12:58] <Extend> i'm currently using mediawiki and i wanna ask about a feature [10:13:08] <VasilievVV> !ask | Extend [10:13:08] --mwbot-- Extend: Don't say "I have a question", or ask "Is anyone around?" or "Can anyone help?". Just ask the question, and someone will help you if they can. Also, please read < http://workaround.org/moin/GettingHelpOnIrc > for a good explanation of getting help on IRC. [10:13:24] <Extend> #if some of you used dokuwiki it has [10:13:26] <Extend> ok [10:13:38] <Extend> in dokuwiki it list the parent of the page [10:13:44] <Extend> how can i do that in mediawiki [10:14:25] <Extend> so i asked ? [10:14:36] <Splarka> enable subpages in main namespace [10:14:47] <CIA-55> 03raymond * r38225 10/trunk/phase3/languages/messages/MessagesEn.php: Add/change a few punctuation marks (.) for consistency with other header/summary messages. [10:15:17] <Extend> how? [10:16:22] <Extend> i will search thanks Slarka [10:18:06] <Jack_Phoenix> Extend, you might wanna have a look at http://www.mediawiki.org/wiki/Manual:%24wgNamespacesWithSubpages for info on how to enabled subpages in the main namespace [10:18:17] <Extend> thanks Jack_Phoenix [10:18:21] <Jack_Phoenix> np [10:20:26] <wikibugs> 03(NEW) Merge messages 'sharedupload' and 'shareduploadwiki'/' shareduploadwiki-desc' - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=14980 normal; normal; MediaWiki: User interface; (jhsoby) [10:32:21] <CIA-55> 03emil * r38226 10/trunk/extensions/TorBlock/TorBlock.class.php: missing global statement [10:34:18] <wikibugs> 03(NEW) Shared repo APIs should include - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14981 15enhancement; normal; MediaWiki: API; (jhsoby) [10:34:45] <Extend> i tried to use this doc http://meta.wikimedia.org/wiki/Subpages [10:34:51] <Extend> but doesn't work for me [10:35:00] <Extend> and this one too http://www.mediawiki.org/wiki/Manual:$wgNamespacesWithSubpages [10:35:08] <Extend> im using mediawiki 1.12 [10:35:11] <Extend> any ideas? [10:35:11] <wikibugs> 14(INVALID) Shared repo APIs should include - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=14981 +comment (10jhsoby) [10:35:13] <Jhs> >< [10:36:23] <Bryan> Extend: define "it doesn't work" [10:37:26] <Extend> the sub pages from the namespace doesn't appear [10:37:29] <Extend> there is no changes [10:37:38] <Extend> i cleared cache ..etc nothing works for me [10:39:36] <Bryan> you tried purgin the page? [10:40:17] <Extend> what do you mean by purging the page ? [10:42:32] <CIA-55> 03raymond * r38227 10/trunk/phase3/ (3 files in 2 dirs): Add text for header of the patrol log for consistency with all other log headers. [10:43:04] <MaxSem> !purge | Extend [10:43:04] --mwbot-- Extend: To purge a cached page, such as when making changes to the navigation bar, add &action=purge to the end of the page url, or ?action=purge if using Simple URLs. E.g: http://en.wikipedia.org/wiki/Main_Page?action=purge [10:44:32] <wikibugs> 04(REOPENED) Shared repo APIs should include 'repo-sharedupload' and ' repo-name' messages - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=14981 summary; +comment (10jhsoby) [10:44:54] <Duesentrieb> actually, to cahnge $wgNamespacesWithSubpages he needs to edit localsettings, which should purge caches automatically [10:45:55] <Extend> MaxSem, i don't have this feature in my mediawiki [10:46:16] <ZorroIII> Jhs: what are you playing with? [10:47:38] <Extend> and the purge doesn't work [10:48:11] <wikibugs> 03(mod) Blocking and tagging from checkuser interface - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=12808 (10mike.lifeguard) [10:48:34] <Duesentrieb> Extend: if you don??t have that feature, you don't have mediawiki [10:48:46] <Extend> wait i will try diffrent [10:48:49] <Extend> way [10:49:32] <Extend> ok works for me [10:49:49] <Extend> but tell me how do i need to purge the entire mediawiki [10:50:15] <Bryan> there should be a maintenance script to invalidate all html caches [10:50:19] <Bryan> let me have a look [10:50:24] <Extend> okay [10:50:30] <Extend> but the feature still [10:50:36] <Extend> i can't get the subpages to work [10:51:39] <Bryan> can't find it [10:52:12] <Extend> and i can't get subpages work ;'( [10:52:36] <jamasi> Extend, are those working in the User: namespace? [10:53:05] <Extend> no [10:53:25] <jamasi> did you do any customisation to the skin? [10:53:45] <Extend> no [10:53:49] <Extend> it's the default [10:54:18] <jamasi> can you link a non-working page here? [10:54:24] <Duesentrieb> Extend: how exactly did you configure $wgNamespacesWithSubpages? [10:54:39] <Extend> # Enable subpages in the main namespace [10:54:40] <Extend> $wgNamespacesWithSubpages[NS_MAIN] = true; [10:54:40] <Duesentrieb> Extend: what namespaces do you want to use subpages with? [10:54:47] <Extend> i want all [10:54:53] <Extend> # Enable subpages in the main namespace [10:54:54] <Extend> $wgNamespacesWithSubpages[NS_MAIN] = true; [10:54:58] <Extend> i set this for all [10:55:02] <Extend> # Enable subpages in all namespaces [10:55:02] <Extend> #$wgNamespacesWithSubpages = array_fill(0, 200, true); [10:55:06] <Extend> but doesn't work too [10:55:17] <Duesentrieb> well you have to remove the # in front [10:55:23] <Duesentrieb> which disables the line [10:55:32] <Duesentrieb> anyway - where in you localsetting did oyu put that? [10:55:32] <Extend> yes i did [10:55:36] <Extend> i just commented it now [10:55:40] <Duesentrieb> it should go near the end [10:55:43] <Extend> i put that at the end [10:55:53] <Duesentrieb> then it should work :) [10:55:59] <Extend> it doesn't work [10:56:09] <Duesentrieb> Extend: [10:56:17] <Duesentrieb> can i have a url of your wiki? [10:58:32] <Extend> yes prvt [11:01:29] <Splarka> also note subpaging is a subtle thing, it only breadcrumbs back in contentSub if the base page exists [11:08:10] <CIA-55> 03ashley * r38228 10/trunk/tools/planet/en/config.ini: add one blog back & fix one url [11:09:50] <TazDaDemon> hey [11:09:52] <TazDaDemon> hi [11:10:20] <TazDaDemon> i've got a question about default mediaWiki settings [11:10:44] <TazDaDemon> i installed it by package manager (apt-get install mediaWiki) [11:11:09] <TazDaDemon> but i don't know where it physicially is installed [11:11:45] <TazDaDemon> can anyone help? [11:13:15] <sentinel100101> well /apt-get/cache is where the deb package would be downloaded use apton(apt-get install aptoncd) to copy from cache and save it externally [11:13:44] <sentinel100101> or do u want to know the path where the file is installed and the rest of the contents are extracted? [11:14:00] <Duesentrieb> dpkg -L mediawiki [11:14:12] <TazDaDemon> i know where the package is, but i dont know where the content of package was installed [11:14:20] <TazDaDemon> yes [11:14:22] <Duesentrieb> see above TazDaDemon [11:14:39] <TazDaDemon> ty [11:14:49] <sentinel100101> as usual.. smart boy ^_^ Duesentrieb [11:14:51] <Duesentrieb> not that distri packages are not really supported by us [11:14:54] <Duesentrieb> and tend to suck badly [11:14:59] <TazDaDemon> umm? [11:15:05] <Duesentrieb> ok, really off for now :) [11:15:06] <Duesentrieb> bbl [11:15:39] <Splarka> !package [11:15:39] --mwbot-- Many Linux distributions provide MediaWiki in a packaged format for that distribution. The MediaWiki development team refers you to your Linux distribution for assistance with installing, configuring or using them. The individual communities & companies who maintain such packages should provide installation instructions. [11:15:43] <Splarka> (the usual response) [11:15:45] <TazDaDemon> instead of those packages, is it suggested to install from sources? [11:16:26] <TazDaDemon> by compiling it [11:16:37] <Splarka> well, it can be hard to find assistance (here) with a package and its unique quirks [11:16:50] <TazDaDemon> i c [11:18:13] <Splarka> http://www.mediawiki.org/wiki/Manual:Installing_MediaWiki [11:29:34] <CIA-55> 03raymond * r38229 10/trunk/phase3/ (4 files in 3 dirs): [11:29:34] <CIA-55> * Add a nice fieldset for consistency [11:29:34] <CIA-55> * Switch result line and input form [11:29:34] <CIA-55> * Do not show result line "Below is a list of 0 files...". We have the message 'noimages' instead. [11:29:36] <MinuteElectron> PHP is interpreted, you do not need to compiler MediaWiki. [11:31:08] <CIA-55> 03raymond * r38230 10/trunk/phase3/CREDITS: Add myself :) [11:33:56] <ThomasOwens> morning. I was wondering if there are any documents specifically related to writing a login extension. I want to be able to allow registered users with the website to be able to (1) carry the session data into the wiki and (2) log in to the wiki if they are not logged into the site and set all site and wiki data there. [11:35:05] <mkv-> Question: I am trying to write a small tag-extension of my own, I have read the tag-extension manual and I cannot find anything wrong in my "script". But it still doesnt work... Any suggestions? Anywone care to look at it? [11:37:26] <mkv-> Instead of returning the values I want it just prints the "tag" that I want it to "hook"... [11:39:08] <mkv-> Whoups, now I found a typo. [11:40:56] <Dantman> mkv-: The developers help pages on MW.org are pretty bad... I suggest taking a look at how some of the more common extensions like ParserFunctions do it [11:41:31] <MaxSem> or in this particular case, CentralAuth [11:41:40] <mkv-> Ok, thanks. [11:42:01] <ThomasOwens> hmm. at one point, someone linked me to documentation specifically related to login extensions and the different variables and such, but I'm not finding that. [11:42:18] <ThomasOwens> MaxSem: if that was to me, CentralAuth is nothing like what I want to do. [11:43:13] <MaxSem> ThomasOwens, it's an example of auth extension [11:43:49] <ThomasOwens> but it's not even close to what I want to do. for one thing, authenticating across wikis is easier because they all use the same username/password system. [11:44:09] <ThomasOwens> we use email address/username/password. [11:44:17] <MaxSem> http://www.mediawiki.org/wiki/Category:User_identity_extensions [11:44:55] <CIA-55> 03siebrand * r38231 10/trunk/extensions/GroupPermissionsManager/GroupPermissionsManager.i18n.php: Add space [11:46:56] <CIA-55> 03siebrand * r38232 10/trunk/extensions/GroupPermissionsManager/GroupPermissionsManager.i18n.php: Revert 'grouppermissions-desc3' back to the revision before Raymond's. Looks like a paste error. [11:48:34] <wikibugs> 03(NEW) Enable patrolling on multilingual Wikisource - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14982 15enhancement; normal; Wikimedia: wikibugs; (yann) [12:04:03] <cirwin> anyone got an idea as to how to change the "Entries in category 'Example'" on a per category basis? [12:04:36] <cirwin> I was going to add a magic word into the category text, but it would appear that the headings are rendered before the text itself [12:06:24] <jamasi> cirwin, you could try to do some hacks in the skin of your wiki [12:06:53] <cirwin> I think I'm going to have to, but how to get the information from the category text into the headings.. [12:08:02] <jamasi> within a skin you can use normal php and you've got access to various MW objects. [12:08:11] <cirwin> yes [12:08:19] <mkv-> If I want to upload a extension to mediawiki's page. How do I do? [12:08:28] <cirwin> mkv-: just edit the wiki pages [12:08:37] <cirwin> if you want it in svn, file a bug report [12:10:44] <mkv-> cirwin: Ah okay [12:20:59] <Dantman> ^_^ Nice... SHID is comming allong [12:21:19] <Dantman> Jack_Phoenix, you might be interested in this one... http://shid.nadir-point.com/user.php?id=1 [12:21:25] <Jhs> ZorroIII, playing with how to improve sharedupload [12:23:58] <^demon> Jhs: Fwiw, ForeignApiRepo isn't a component of the API. :-) [12:24:10] <^demon> (was just looking at bug 14981) [12:25:45] <Jhs> ^demon, it isn't? what's it part of then [12:25:52] <^demon> The filerepo code. [12:26:05] <^demon> It _uses_ the remote API, but it's not really a part of it. [12:26:14] <adinc> hello, i moved the directory of my wiki installation and now the design is messed up. do i have to consider something particular? [12:26:15] <Jhs> okay, i see [12:26:36] <Jhs> adinc, url? [12:26:44] <adinc> Jhs: yes the url changed aswell [12:26:56] <adinc> it was before a subdirectory and now i did a subdomain for it [12:27:01] <Jhs> .... [12:27:06] <adinc> ahhh, it is wiki.dinc.org [12:27:09] <Jhs> :D [12:27:14] <adinc> ;-) [12:28:06] <Jhs> i assume you have to change the directory path in the config somewhere [12:28:38] <adinc> jhs yes, thats what i thought aswell, but the LocalSettings file doesnt have any directory parameter [12:29:03] <Jhs> hmm... [12:29:29] <^demon> adinc: Check your local settings and make sure $wgScriptPath is set correctly. In your case, it looks like it needs to be set to / [12:29:47] <Duesentrieb> which is evil [12:29:55] <Duesentrieb> don't install mediawiki files into the document root [12:29:58] <Duesentrieb> that spells trouble [12:29:58] <adinc> ^demon: ohhh yes, sure [12:30:01] <adinc> thank you [12:30:53] <CIA-55> 03aaron * r38233 10/trunk/extensions/FlaggedRevs/ (5 files): Refactor some code to look better [12:30:57] <adinc> ^demon: ohh, i just see that it didnt help [12:32:08] <Werdna> that's annoying [12:32:17] <Werdna> every '|' seems to evaluate to true [12:34:31] <ThomasOwens> Where does MediaWiki normalize the user name provided on Special:Userlogin? My custom auth is almost done, but I think it's rejecting "user names" because we use email addresses for login purposes. [12:34:44] <Werdna> it's not normalisation [12:34:50] <vinodhrajan> Hi. I Enabled Sub pages in the Main Namespace. However Subpages appear in some pages and do not appear in other pages.. [12:34:58] <Werdna> but you are quite correct that mediawiki doesn't like '@' in usernames [12:35:15] <ThomasOwens> See, all data is in an existing user database. [12:35:21] <ThomasOwens> So what are my options? [12:35:55] <ThomasOwens> Even if it involves hacking the code behind Special:Userlogin, as I'm looking at that now. [12:36:40] <vinodhrajan> see: http://www.tamilbuddhism.com/wiki/%E0%AE%AE%E0%AF%81%E0%AE%A4%E0%AE%B1%E0%AF%8D_%E0%AE%AA%E0%AE%95%E0%AF%8D%E0%AE%95%E0%AE%AE%E0%AF%8D/demo & http://www.tamilbuddhism.com/wiki/%E0%AE%AA%E0%AF%8C%E0%AE%A4%E0%AF%8D%E0%AE%A4%E0%AE%AE%E0%AF%81%E0%AE%AE%E0%AF%8D_%E0%AE%A4%E0%AE%AE%E0%AE%BF%E0%AE%B4%E0%AF%81%E0%AE%AE%E0%AF%8D/%E0%AE%AE%E0%AF%81%E0%AE%A9%E0%AF%8D%E0%AE%A9%E0%AF%81%E0%AE%B0%E0%AF%88 [12:37:06] <vinodhrajan> Subpages appears in the First one but doesn't appear in the seconde one :( [12:37:20] <Werdna> ThomasOwens: well, it could be done, but it interferes with some stuff [12:37:31] <Werdna> we used to allow '@' [12:37:31] <ThomasOwens> could you elaborate? [12:37:56] <Werdna> but, for instance, changing user rights cross-wiki requires @-free usernames [12:38:02] <cirwin> there are users on wiktionary with an @ [12:38:13] <ThomasOwens> ok, well we only have 1 wiki, so that's not a problem. [12:38:13] <Werdna> yes, legacy [12:38:19] <Werdna> we don't allow new registrations with '@' [12:38:34] <cirwin> bad coding if you use a legal char as a seperator ;p [12:39:10] <ThomasOwens> So what do I need to do to allow @ in user names? Because that's the only thing stopping me from using our current database.. [12:39:15] <Werdna> ThomasOwens: see static function isUsableName( $name ) { [12:39:19] <Werdna> in User.php [12:39:23] <Werdna> oh wait [12:39:25] <Werdna> the one below it [12:39:28] <Werdna> static function isCreatableName( $name ) { [12:39:33] <ThomasOwens> ok. thanks. [12:39:40] <Werdna> your own risk :P [12:40:31] <ThomasOwens> I know. [12:40:43] <Jhs> ThomasOwens, my suggestion is that you replace the @ with (a) or something in your database, and write a note on special:userlogin that users with @ in their names have to use (a) instead [12:41:18] <^demon> adinc: Not sure :( [12:41:24] <Jhs> only noobs want @s in their names anyways [12:41:34] <adinc> ^demon: ;-) [12:43:38] <adinc> i've set up login for my wiki, and when i call the mail page an erro page appears. can i get a loginpage to appear directly? [12:44:24] <wikibugs> 03(mod) $wgServer lacks brackets in IPv6 URLs - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14977 (10gillou.ray) [12:50:13] <ThomasOwens> hmm. [12:50:43] <ThomasOwens> is there something special I need to do for an auth extension, other than include it and create a new instance of it in LocalSettings? [12:50:51] <ThomasOwens> unless... [12:55:38] <SeeeD> hey, i wan't to create a mainpage like wikipedia... which plugins does i need? [12:55:51] <SeeeD> Can anybody say a plugin? :) [12:56:57] <Jhs> SeeeD, 0 [12:57:39] <Jhs> all you need is already there [12:57:48] <Jhs> (software-wise) [12:57:56] <Jhs> you will probably need to copy some css though [13:01:15] <SeeeD> hm? [13:01:25] <SeeeD> does i have to search for (software-wise)? [13:02:01] <SeeeD> ah [13:02:02] <SeeeD> no [13:02:03] <SeeeD> mh [13:02:09] <SeeeD> is there any tutorial for this? [13:02:26] <SeeeD> ahh [13:02:27] <SeeeD> stop [13:02:34] <SeeeD> i have to copy the css from wikipedia :D [13:02:35] <SeeeD> w8 [13:02:40] <SeeeD> i try it ;) [13:09:29] <SeeeD> is there no plugin, which lists the newest artikel of any category? [13:10:02] <cirwin> SeeeD: http://www.mediawiki.org/wiki/Extension:DynamicPageList [13:10:13] <SeeeD> thx :) [13:11:16] <ThomasOwens> ok. so with my auth extension, I can "log in", but it always says that you are logged in as an IP address and you aren't really logged in (you have the "log in/register" link instead of the logout link). [13:11:19] <ThomasOwens> thoughts? [13:12:26] <CIA-55> 03shinjiman * r38234 10/trunk/phase3/languages/messages/ (4 files): Localisation updates Cantonese, Chinese and Old/Late Time Chinese [13:13:15] <wikibugs> 03(mod) Get undefined variable:sid errors after upgrade to SMW 1.2 - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14812 (10mitchell_neill) [13:14:38] <Nikerabbit> hello [13:14:58] <ThomasOwens> wait. I might have fixed my problem. I hate typos. [13:20:51] <CIA-55> 03raymond * r38235 10/trunk/phase3/includes/ImagePage.php: Remove unused var $dims [13:32:04] <alpha_q> hey [13:32:27] <alpha_q> how can use I the current user name in a template [13:32:40] <MinuteElectron> not possible [13:32:41] <alpha_q> ~~~~ just replaces my name on save [13:33:05] <MinuteElectron> right [13:33:13] <alpha_q> well not the current, but the user name with which the template was inserted [13:33:36] <Duesentrieb> alpha_q: that can be done by providing ~~~ as a template param [13:33:58] <Duesentrieb> it will be replaced on save, so the param will be whover placed the template [13:34:04] <Duesentrieb> there is no way to fully automate this [13:34:18] <alpha_q> ok (so we were talking about the same thing) [13:34:19] <alpha_q> thanks [13:34:58] <Duesentrieb> (well, if you subst the template itself, there are some tricks iirc. but it doesn't help really) [13:35:03] <ThomasOwens> is there a list of the session variables that are used? [13:35:40] <Duesentrieb> ThomasOwens: don't know, but you can print_r($_SESSION);... [13:36:11] <alpha_q> hey [13:36:26] <alpha_q> a what about ~<includeonly>~</includeonly>~~ ? [13:36:30] <alpha_q> won't this do the trick? [13:36:40] <Duesentrieb> it will show as literal ~~~ [13:36:44] <Duesentrieb> unless you subst the template [13:37:00] <Duesentrieb> at least, i think so [13:37:02] <Duesentrieb> try for yourself :) [13:37:51] <Duesentrieb> alpha_q: btw, three ~~~give just the user name, four include a date (five give just the date... which is a misfeature, really) [13:38:40] <alpha_q> yeah I tried that :) [13:38:50] <alpha_q> just displayed ~~~~ on the page [13:38:59] <alpha_q> well I guess that's not such a big deal anyway [13:39:02] <alpha_q> thanks for the help [13:39:56] <Duesentrieb> {{foo|~~~}} does the trick, even though it's a bit nasty to type [13:40:09] <alpha_q> yeah that's why I needed it :) [13:41:31] <Duesentrieb> if you always need it for the same template, you can make a button for it on the edit page [13:41:38] <Duesentrieb> !editbuttons [13:41:38] --mwbot-- See <http://www.mediawiki.org/wiki/Manual:FAQ#How_do_I_add_more_buttons_on_the_edit_page.3F>, also have a look at <http://www.mediawiki.org/wiki/Extension:CharInsert> [13:59:08] <CIA-55> 03jhsoby * r38236 10/trunk/ (2 files in 2 dirs): Changing User:E's blog link from search to feed, so it works properly [13:59:26] <Xiong> savage criticism requested: [13:59:29] <Xiong> http://www.beyondeuclid.com/wiki/Proof%3ACircumscribed_Trapezoid [14:05:27] <Jhs> lol @ brion's comment here: https://bugzilla.wikimedia.org/show_bug.cgi?id=13784 [14:05:32] <CIA-55> 03raymond * r38237 10/trunk/phase3/languages/messages/MessagesEn.php: Partly revert of r38236. looks like a commit accident by Jhs [14:09:14] <CIA-55> 03demon * r38238 10/trunk/phase3/includes/parser/CoreParserFunctions.php: I love case-sensitivity. Don't you? [14:09:48] <^demon|away> Jhs: I spent like a few hours tracking down my friend trying to get him to translate in hopes of it actually being useful. [14:10:33] <Jhs> ^demon|away, very good. into what language? [14:10:36] <Jhs> ah [14:10:44] <VasilVV> ^demon|away: Google translate? [14:10:45] <Jhs> about brion's comment [14:10:52] <Jhs> i forgot i posted that above :P [14:10:55] <^demon|away> VasilVV: No, native speaker :-P [14:12:47] <Xiong> i despise case sensitivity [14:13:23] <Xiong> i think it is (a) useless and (b) contrary to the semantic value of the letters of the alphabet [14:13:36] <Xiong> as commonly understood [14:13:51] <Xiong> A and a do not carry distinct semantic payloads [14:14:21] <Nikerabbit> Xiong: maybe, but Apple is very different from apple [14:14:26] <Xiong> case, rather, imputes an overtone to meaning -- it is not basic [14:14:27] <Nikerabbit> it is also slower [14:14:32] <Xiong> i disagree [14:14:54] <Xiong> standing alone, neither Apple nor apple uniquely distingish [14:14:56] <CIA-55> 03jhsoby * r38239 10/trunk/phase3/languages/messages/MessagesEn.php: Fixing capitalisation in two English messages [14:15:13] <Xiong> it is lazy and arrogant to demand of the reader that he infer your meaning [14:15:16] <^demon|away> Jhs: Capitalization fix day? [14:15:25] <Nikerabbit> in proper writing Apple can refer to things apple cannot [14:15:26] <Xiong> rather, say apple the fruit or apple computer company [14:15:46] <Jhs> ^demon|away, (not so away :P ) indeed [14:15:51] <^demon|away> Personally, I would love it if PHP knew what I _meant_ instead of what I _typed_. Would make writing code a million times easier :-P [14:15:54] <^demon> Oops [14:15:55] <Xiong> this is not about the manual of style -- any manual of style [14:15:58] <VasilVV> Xiong: apple if fruit, but Apple is a company [14:16:01] <Xiong> this is about clear communication [14:16:08] <Jhs> i think i killed most of it earlier though, but i found some things this week as well [14:16:33] <Xiong> this is about assuming the burden of communication when one is writing [14:16:41] <Xiong> and not placing it on the shoulders of the reader [14:17:11] <Nikerabbit> Xiong: that's not how communication works [14:17:16] <Xiong> this includes article text, irc chat, search boxes, and computer code [14:17:25] <Xiong> i say, it is [14:17:45] <^demon> Jhs: I was looking at a capitalization bug like 2 minutes ago, I cannot find it now for the life of me. [14:17:46] <Xiong> the primary burden of clear communication rests on the sender [14:18:09] <Jhs> ^demon, hehe [14:18:12] <Nikerabbit> people tend to choose the shortest expression that can still be understood without big problems [14:18:15] <^demon> Jhs: Bug 14800. [14:18:24] <Xiong> if you mean to communicate a fact, say, "apple computer's second quarter was profitable" [14:19:12] <Xiong> then it is not necessary to lard this with an overtone that suggests the sender is somehow more sophisticated because he says, by a strict application of rules, he can disambiguate with case [14:19:33] <Xiong> we live in a world where some people still cannot read at all [14:19:51] <Xiong> and where almost everyone is taught that there are 26 letters in the alphabet, not 52 [14:19:54] <Nikerabbit> yeah, and now we can end this discussion [14:20:04] <Jhs> ^demon, ah yes. i saw that one earlier [14:20:05] <Xiong> because you're right? [14:20:22] <Nikerabbit> Xiong: no, because it doesn't lead to anything [14:20:28] <Xiong> disagree [14:20:38] <Xiong> it leads directly to implementation of case insensitivity [14:20:48] <Xiong> which i have wanted for a very long time [14:21:04] <Nikerabbit> Xiong: and note that most languages do not even have alphabet with big and small letters... let alone letters [14:21:22] <Xiong> i'm only concerned with english [14:21:24] <Jhs> ^demon, i might actually have fixed that one unknowingly, with r38207 [14:21:36] <Xiong> case sensitivity in MW is an unnecessary complication [14:21:39] <Jhs> but i don't know if the log uses group-* or group-*-member [14:21:50] <Xiong> it leads to thousands of time-consuming redirects, on WP [14:22:16] <Nikerabbit> Xiong: now you are changing topics [14:22:21] <^demon> Jhs: Go for it ;-) [14:22:30] <Xiong> and on other installs, requires careful attention to case when linking, *if* one chooses to follow correct style by capitalizing the "counter" words of titles [14:22:45] <Xiong> now i'm confused [14:22:52] <Xiong> i thought i was discussing case sensitivity [14:23:06] <Xiong> my thesis is, it is a Bad Thing [14:23:16] <Xiong> i don't say that case itself is wrong [14:23:19] *^demon totters off to fix a case-sensitivity bug [14:23:19] <Nikerabbit> if you want case insensitive titles, go file a bug (if there isn't one) or provide a patch [14:23:25] <Xiong> i believe in case preservation [14:23:30] <Nikerabbit> it only affects 10% languages or so anyway [14:23:36] <Xiong> i'm told the demand has been made before [14:23:40] <domas> case insensitive titles are easy [14:23:44] <^demon> Nikerabbit: I think there is one....probably more than one. [14:23:47] <domas> just use case insensitive collation [14:23:48] <domas> :) [14:23:57] <Xiong> i've even been told that case sensitivity is scattered through the code, requiring much work to eradicate [14:23:58] <^demon> hi domas :-) [14:24:04] <domas> HI CHAD [14:24:19] <domas> Xiong: not really, if database is case insensitive, there's not much done at mediawiki layer [14:24:20] <Xiong> well domas, you are the first person to suggest an easy fix [14:24:23] <^demon> Xiong: Case sensitivity of first character is controlled through $wgCapitalLinks. [14:24:29] <Xiong> would you like to go into a bit more detail? [14:24:36] <Xiong> demon, i'm aware of that switch [14:24:50] <domas> Xiong: ALTER TABLE ... CONVERT TO CHARACTER SET ... COLLATE ... [14:24:53] <Xiong> it only makes things more case sensitive, though, not less [14:25:01] <domas> Xiong: for most of tables. change varbinary fields into varchar, probably :) [14:25:05] <Xiong> domas, i don't follow you [14:25:11] <domas> Xiong: thats your problem then! :) [14:25:20] <Xiong> no need to be abrupt, sir [14:25:32] <Xiong> you seem to have a solution and i dearly would like to be able to apply it [14:25:36] <Xiong> consider me ignorant [14:26:26] <Xiong> do you have a method for making links completely case insensitive, while preserving case? [14:27:15] <Xiong> if so, and if the procedure is involved, as it appears, i'd like to ask you very nicely to write it up and post it on MW somewhere [14:27:24] <Xiong> i will give you a cookie for it :) [14:32:24] <sentinel100101> hah [14:32:25] <sentinel100101> haha [14:32:36] <sentinel100101> thats was kinda funny ^_^ [14:32:46] <sentinel100101> no offence [14:32:47] <sentinel100101> peace [14:32:49] <Dantman> ^_^ Pay me a wage and I'll get on the titlerewite [14:33:12] <Xiong> if i had money, i would [14:33:16] <Dantman> heh [14:33:41] <Xiong> i may feel more strongly about other issues but this one at least is fixable [14:33:51] <Xiong> no cure for cancer, no hope for world peace [14:35:31] <^demon> Dantman: We could trade features. You do Title rewrite and I'll do blocking rewrite :-P [14:36:00] <Dantman> heh... Title rewrite was mine... [14:36:14] <Dantman> But I need money if I'm going to work on something nowadays [14:36:21] <^demon> domas: Mind looking at a patch on bug 14418? It's a schema change I'm looking to do. [14:36:38] <domas> :( [14:36:49] <Dantman> I can't afford to work on something anymore unless it either gives me money, or something to put in a portfolio... [14:37:01] <^demon> Minor change <3 [14:37:13] <Dantman> And nothing MW related fits in a usable portfolio for me [14:37:55] <domas> Dantman: neither in mine! [14:37:56] <domas> :( [14:38:24] <domas> ^demon: I strangle you for char() alone [14:38:31] <Xiong> dantman that is what i tell my clients [14:38:40] <domas> ^demon: should I continue reading? [14:38:56] <Xiong> i told that to one guy and he promptly hired a manager who is now trying to kill my project entirely [14:39:05] <^demon> Yes, I justified my use of char(32), for it is also used in the interwiki table for the iw_prefix column. [14:39:20] <domas> ^demon: why not varchar() ? [14:39:41] <Xiong> that'll teach me to ask for resources [14:39:47] <^demon> domas: No particular reason, just stole the column format from the interwiki table. [14:39:59] <domas> ^demon: then fix it in interwiki table too ;-) [14:40:11] <domas> actually we don't use interwiki table on live site [14:40:59] <domas> but by putting that stuff into redirects table [14:41:05] <domas> you really deserve some punishment [14:41:23] <Schroeder> I wonder what propaganda Kremlin tool RIA Novosti is putting out today... [14:41:25] <^demon> uh oh :( [14:41:31] <Simetrical> domas, doesn't MySQL secretly convert long char to varchar anyway? [14:41:39] <domas> Simetrical: no [14:42:06] <domas> ^demon: another problem, you realize that without proper indexing that addition won't be usable in any way ? :) [14:42:28] <^demon> I really suck at SQL, that much is becoming more obvious :-P [14:42:53] <domas> thats the usual problem with mediawiki development [14:43:06] <domas> people add columns without thinking what they will be used for [14:43:25] <domas> your patch just fills the data [14:43:28] <domas> so it is entirely useless [14:43:31] <domas> as it doesn't use it [14:43:34] <domas> :) [14:43:58] <Simetrical> domas, it doesn't need to be indexed, unless you want a list of all redirects pointing to a specific wiki. [14:44:14] <domas> Simetrical: why does that field matter, then? [14:44:27] <^demon> Simetrical: Which would be cool. As would being able to remove interwiki redirects from the "Broken Redirects" page. [14:44:42] <Simetrical> The point would be so that you could filter out interwiki redirects from lists for some special pages, for instance. You *could* make the index bigger for that, of course. [14:44:45] <Simetrical> Or add another index. [14:45:08] <Simetrical> But since almost all redirects are intrawiki it might be fastest to not have to maintain the index, yes? [14:45:25] <domas> in theory you can prepend the key with interwiki, blank being 'not interwiki' [14:45:29] <domas> but it would be a hack [14:45:32] <domas> and would suck anyway :) [14:45:49] <domas> in the end, we don't really know what we're doing with that data anyway [14:45:52] <domas> so why store it there? [14:45:53] <^demon> domas: It currently saves the title with the iw prefix attached. [14:46:38] <Dantman> eek... [14:46:42] <domas> I really have troubles thinking about something [14:46:48] <domas> where you don't yourself know what you're doing [14:46:51] <domas> why should I know what you're doing? [14:46:58] <Simetrical> domas, by the way, I think we need ug_sortkey pretty soon. http://en.wikipedia.org/w/index.php?title=Special%3AListUsers&username=&group=rollbacker&limit=50 [14:47:36] <domas> gh [14:47:45] <Nikerabbit> domas: are you offended if someone says that Lithuanian is a tiny language? [14:48:22] <domas> Nikerabbit: offended? no. do I think the one who says that is clueless? yes :) [14:48:25] <^demon> domas: It has potential for uses, such as specialpage filtering, etc by IW prefix. Also a blocker for a bug with Special:BrokenRedirects. [14:48:26] <wikibugs> 03(NEW) Patch to make Special: Statistics result rendering consistent with Special:Search - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14983 normal; normal; MediaWiki: Special pages; (hightowe) [14:48:37] <domas> Simetrical: are we giving out groups automatically nowadays? [14:48:46] <Nikerabbit> domas: hehe, I wouldn't consider it or Welsh very tiny either [14:48:54] <domas> ^demon: I don't care about features that have potential for uses ;-) [14:48:56] <Simetrical> domas, no, just to anyone who asks and isn't obviously a vandal. :P [14:48:57] <Jhs> domas, yes, we are; at least global groups [14:49:14] <Jhs> n/m [14:50:25] <^demon> domas: Fair enough. :-P Wontfix? Clean it up a bunch and find a really good reason to use it then bug you again? [14:50:58] <domas> ^demon: the only thing I care there about is CHAR() [14:51:05] <domas> ^demon: it is wrong wrong wrong [14:51:10] <^demon> Easily fixed. [14:51:20] <^demon> I'll play with that later. [14:51:32] <domas> anyone who writes updaters removing all CHAR()s and TINYBLOBs from mediawiki codebase [14:51:36] <domas> and replacing them with VARCHARs [14:51:38] <domas> will get a cookie [14:51:42] <domas> I'm doing that on live site anyway [14:52:00] <Simetrical> What if something is actually fixed-length? [14:52:04] <Simetrical> Like img_sha1? [14:52:55] <domas> would work then [14:53:08] <domas> sucks though when we don't use NULLs for empty values [14:53:21] <Simetrical> There are no empty values in img_sha1, in theory. [14:53:29] <domas> if you have populated properly [14:53:39] <Dantman> eh? I thought we were trying to move away from text types and into binary types [14:53:59] <Simetrical> Dantman, char = binary in MySQL 4, which is all Domas cares about. [14:54:08] <Simetrical> No character collations until 4.1 or whatever. [14:54:24] <domas> Dantman: not an issue, 'CREATE DATABASE .. CHARACTER SET = binary' [14:54:29] <domas> and all your varchars become varbinary [14:54:35] <domas> the problem is with CHAR() though [14:54:37] <Simetrical> The point is var vs. non-var, and char/binary vs. tinytext/tinyblob. [14:54:42] <domas> because then it becomes BINARY() with zero-byte padding [14:54:53] <domas> and PHP hates zero-byte padding [14:55:17] <wikibugs> 03(NEW) Moving a page does not trigger autoflagging - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14984 normal; normal; MediaWiki extensions: FlaggedRevs; (pbirken) [14:55:44] <Dantman> I recall the manual saying something about VARCHAR BINARY not being treated the same as something that's actually binary [14:56:30] <domas> there's a difference between 'varchar binary' and 'varchar character set binary' [14:56:37] <domas> 'varchar binary' just picks binary collation [14:56:51] <domas> 'character set binary' makes it varbinary [14:56:59] <domas> a mess, but so it goes :) [15:09:50] <Xiong> BASEPAGENAME does not work as advertised [15:10:28] <Xiong> it seems that {{BASEPAGENAME}} on the page Foo:Bar/2 resolves to Bar/2, not Bar [15:10:46] <Duesentrieb> Xiong: got subpages enabled in that NS? [15:10:59] <Xiong> yabba yabba uh [15:11:03] <Xiong> blort [15:11:07] <Duesentrieb> :P [15:11:07] <Xiong> dunno [15:11:36] <Duesentrieb> well, if youdo, there would be a link to Foo:Bar on top of Foo:Bar/2 [15:11:39] <Xiong> probably not, it's an extra [15:11:51] <Duesentrieb> indeed [15:12:12] <Xiong> i need to go an enable subpages on all the namespaces [15:12:25] <Xiong> i can't think of a reason *not* to have subpages [15:12:38] <Duesentrieb> $wgNamespacesWithSubpages = array_fill(0, 200, true); [15:12:53] <Duesentrieb> a good reason is having valid literal names that contain a / [15:13:02] <Duesentrieb> which may be the case especially in the main namespace [15:13:13] <Xiong> ah, that's much better than what i feared, with a giant list of individual things [15:13:19] <Xiong> in my case, never [15:13:23] <Xiong> no fear there [15:13:30] <Duesentrieb> go ahead then [15:13:35] <Xiong> i abhor special characters in titles [15:13:47] <Duesentrieb> "special" is relative [15:13:59] <Duesentrieb> and the Wikipedia article about OS/2 should have *that* title [15:14:17] <Xiong> well, there ya go [15:14:18] <Duesentrieb> some for C++ [15:14:21] <Xiong> i don't like anybody [15:14:22] <Duesentrieb> or A* search [15:14:43] <Xiong> this drive to name things with special characters [15:15:19] <Xiong> i grew up believing identifiers should be limited strictly to A-Z, 0-9, and by the grace of the ghost that used to be a bug in the machine, the underbar [15:15:20] <Duesentrieb> well, the idea is to name things "naturally" and not restrict your self due to some artificial technicality [15:15:46] <cfedde> I'm having trouble with xcache and mediawiki I get the error: "xcache.var_size is either 0 or too small to enable var data caching" The normal google-search-for-error-text technique is not leading me to a solution. Does anyone know where I can find a clue about fixing this? [15:15:52] <Duesentrieb> yes... well, for internal identifiers, i agreee with that, though java allows you to use sanscrit identifiers :) [15:15:54] <Xiong> you're thinking of just the wiki -- i'm on a jihad against the world [15:15:56] <Duesentrieb> but wiki pages... [15:16:01] <Xiong> starting with wal*mart [15:16:23] <Xiong> and the artist formerly known as prince [15:16:33] <Duesentrieb> cfedde: perhaps look into the xchache documentation for info on setting var_size ? [15:16:35] <Xiong> does he have his own private unicode point now? [15:16:56] <Duesentrieb> Xiong: possible :) [15:17:12] <Xiong> i'm going to edit the MW docs to include your shotgun solution; i like it [15:17:28] <Duesentrieb> Xiong: ??u?????? 's?????? ??noq?? ??o?? [15:17:55] <ZorroIII> funny [15:19:01] <Duesentrieb> istole that solution somewhre [15:20:34] <Xiong> how can i point out that '200' suggests 200 ns without being snarky? [15:20:57] <Xiong> "of course, if you have more than 200 ns, you may need to increase that limit" [15:21:11] <cfedde> I suppose that I could have mentioned that I've also been looking in the doc. Thanks for the reminder though. [15:22:16] <cfedde> apparently php and I have a disagreement on where the xcache.ini file is. [15:22:26] <Duesentrieb> cfedde: so you tried chaning it in the [xcache] section in sour php ini by setting something like xcache.var_size = 8M ? [15:22:30] <Xiong> guess who is going to win that argument? [15:22:55] <Duesentrieb> cfedde: well, from what google tells me, it uses php.ini. but that may be wrong or depend on the install. i never used it. [15:23:34] <Nikerabbit> 200 nanoseconds? [15:23:42] <Duesentrieb> .oO(best way to find out wether a config file is actually used: break the syntax) [15:24:09] <cfedde> Duesentrieb: that's what I thought too. this ubuntu has an "interesting directory structure" including directories and such. maybe it's more an ubuntu question. [15:24:10] <Nikerabbit> who was just talking about "say what you mean" and don't let the reader guess... [15:24:13] <cfedde> But thanks for the hints [15:25:06] <Duesentrieb> cfedde: on my ubuntu, its at /etc/php5/apache2/php.ini iirc [15:25:11] <Duesentrieb> no funny directories [15:25:29] <Nikerabbit> Duesentrieb: fcgi <3 [15:25:45] <Xiong> am i correct in thinking that if i enable subpages, having already created a /foo page, that some godawful thing will happen akin to creating a new namespace over existing pages? [15:26:10] <Duesentrieb> Xiong: no, it'll just work [15:26:22] <Duesentrieb> Xiong: just purge caches. which happens automatically wheny ou edit the config [15:26:23] <Xiong> that, i don't understand [15:26:26] <cfedde> ls /etc/php5/ -> apache2/ cli/ conf.d/ [15:26:40] <Duesentrieb> Xiong: the reason is that namespaces are stored in a se??parate database field, while subpages are not. [15:27:00] <Duesentrieb> Xiong: namespaces are actually represented in the database, subpages are jjust a hack with special parsing of page names [15:27:06] <Xiong> well it's a Good Thing anyway [15:27:26] <Duesentrieb> well, it's convenient. if it's Good is another question. [15:27:37] <Duesentrieb> it makes it rather expensive to find all subpages of a given page, for one thing [15:27:50] <Xiong> i do get the backlink to the base page now, when reloading the (empty) subpage [15:28:05] <Xiong> so the gimmick works, thank you much [15:28:35] <Xiong> i'd noticed the enable subpage stuff before but back burnered it, unwilling to write out for all my extras [15:28:39] <Xiong> since i have several [15:29:02] <Xiong> i guess i figured i'd ignore the issue until it bit me on the butt [15:29:05] <Xiong> which it did [15:43:37] <Xiong> okay, i have beaten this punk to a pulp [15:43:39] <Xiong> http://www.beyondeuclid.com/wiki/Circumscribed_Trapezoid [15:43:52] <DanielTahar> hi [15:43:54] <Xiong> i really would appreciate someone telling me if i'm going all wrong [15:44:07] <Xiong> before i continue to go all wrong [15:44:16] <Xiong> hi DT welcome to MW [15:44:18] <DanielTahar> quick question: is there a parameter that translates into the page name? i.e. "you are currently vieweing the "<<PAGENAME>>" page" [15:44:25] <Xiong> yes [15:44:31] <Xiong> {{PAGENAME}} [15:44:42] <DanielTahar> really? well i was very close i see [15:44:48] <DanielTahar> that was a wild guess on my part [15:44:50] <Xiong> maybe maybe not [15:45:03] <Xiong> http://en.wikipedia.org/wiki/Help%3AMagic_words#Page_names_and_related_info [15:45:13] <Xiong> there's more than one choice [15:45:48] <Xiong> note that for all the good stuff, you need to upgrade your install, if it's old [15:49:55] <Xiong> maybe you, DT, would be kind enough to check out my thing and tell me if i've done anything that looks likely to catch fire and burn out of control: [15:50:03] <Xiong> http://www.beyondeuclid.com/wiki/Circumscribed_Trapezoid [16:01:35] <CIA-55> 03demon * r38240 10/trunk/extensions/StalkerLog/ (. README StalkerLog.i18n.php StalkerLog.php): Add StalkerLog extension I wrote up for work awhile back (but we never ended up using). Despite the name, it's targeted at corporate wikis for office productivity monitoring, etc. [16:02:03] <proletaire> Hi everybody [16:02:45] <CIA-55> 03demon * r38241 10/trunk/extensions/StalkerLog/README: svn:eol-style. [16:03:40] <proletaire> Can someone help me to install the FlaggedRevs extension to my mediawiki? [16:04:17] <proletaire> I'm a huge newbie [16:04:51] <yannf> can someone tell me which doc is ok: http://www.visolve.com/squid/whitepapers/reverseproxy.php or http://wiki.squid-cache.org/SquidFaq/ReverseProxy [16:05:12] <yannf> with the 2nd, squid doesn't even start [16:06:24] <cirwin> proletaire: do you know how to use svn? [16:06:29] <proletaire> Really? No good Samaritan? [16:06:36] <proletaire> sry [16:06:42] <guyvdb> Is it more efficient to iterate through part of an array with next() than with a for loop? [16:06:43] <proletaire> no i don't [16:06:48] <cirwin> ok [16:06:58] <cirwin> do you have access to your servers command line? [16:07:27] <Nikerabbit> guyvdb: why would it be? [16:07:30] <proletaire> heum... I tryed connecting with Putty, doesnt work in SSH, only in FTP [16:07:35] <cirwin> ok [16:07:42] <proletaire> does that answer the question? [16:07:45] <cirwin> in that case you need to download all the files from mediawiki [16:07:49] <cirwin> yes, it does [16:08:02] <proletaire> I did download the extension [16:08:07] <cirwin> ok [16:08:11] <proletaire> and uploaded it [16:08:12] <cirwin> then you need to upload it [16:08:14] <cirwin> good :) [16:08:16] <proletaire> of course [16:08:18] <MinuteElectron> guyvdb: You could do a benchmark if you're microoptimising. ;) [16:08:22] <cirwin> so you don't need too much help then :p [16:08:26] <cirwin> where did you put it? [16:08:32] <proletaire> It gives me a PHP error [16:08:38] <cirwin> ok [16:08:41] <cirwin> which php error? [16:08:42] <proletaire> wait [16:09:14] <guyvdb> Nikerabbit: because there might be a table lookup less? [16:09:30] <Nikerabbit> guyvdb: table lookup? [16:10:08] <Nikerabbit> last time I checked next() is a function... [16:10:20] <proletaire> Fatal error: Call to undefined method OutputPage::appendSubtitle() [16:10:32] <proletaire> I have version 1.12 of mediawiki [16:10:44] <proletaire> maybe its the version? [16:10:46] <cirwin> yes [16:10:57] <cirwin> FlaggedRevs needs 1.13 (r37158+) [16:11:02] <proletaire> oh............ [16:11:03] <proletaire> ok [16:11:04] <cirwin> it says so on the extension page [16:11:13] <proletaire> heu [16:11:21] <proletaire> how do I update? [16:11:27] <cirwin> with svn :) [16:11:42] <proletaire> Ok... well lets talk more about SVN then [16:12:00] <proletaire> what is it and how does it work? [16:12:23] <cirwin> it is just a way download lots of files and keep them up to date with minimal effotr [16:12:29] <cirwin> if you can't use ssh onto the server [16:12:35] <cirwin> then you have to use svn on your local computer [16:12:48] <cirwin> for which you need to download and install TortoiseSVN [16:12:59] <cirwin> then once you have downloaded the files, use ftp to upload them all [16:13:03] <cirwin> but it will take you a while [16:13:30] <proletaire> Why? I can't be no more than 20Mo... [16:13:53] <proletaire> And why can't I just download the update and reupload it on my server? [16:14:02] <cirwin> you can [16:14:09] <cirwin> if there is a download for a high enough revision [16:14:18] <^demon> Cirwin: 1.13 rc1? [16:14:53] <cirwin> dunno, what revision is that? [16:15:02] <cirwin> > 37158? [16:15:09] <^demon> Um, should be. [16:15:11] <cirwin> ok [16:15:14] <cirwin> do that then [16:15:18] <proletaire> You have a link? [16:15:48] <^demon> Yeah, 1.13 was branched at 37957. [16:16:37] <^demon> http://download.wikimedia.org/mediawiki/1.13/mediawiki-1.13.0rc1.tar.gz [16:17:45] <proletaire> Ok, but I guess, since I don't use that SVN thing, that I will have to replace to whole mediawiki on my server and install this new version [16:17:47] <proletaire> ? [16:17:55] <proletaire> Thank you btw [16:18:05] <cirwin> you will have to overwrite all the files [16:18:09] <cirwin> not sure about the database [16:18:14] <barcalex> hi, i need your help. i have updated my wiki to the newest version. but now when i want to edit/write an article the editor bar has gone. can someone help ? [16:18:24] <Xiong> 1.13 is now the official, not the bleeding edge? [16:18:35] <proletaire> Hmmm... not really intelligent for futures updates... [16:18:35] <^demon> If you can't run update.php from the command line, try doing a fresh install (AFTER BACKING UP YOUR LOCALSETTINGS FILE!) [16:18:53] <^demon> Xiong: 1.13 has been branched and is in RC status. 1.14alpha is trunk. [16:19:14] <Xiong> don't parse RC [16:19:20] <^demon> Release candidate [16:19:25] <Xiong> ah [16:19:27] <^demon> Anyway, I'm off. [16:19:31] <Xiong> then i will stay away [16:19:55] <proletaire> Do you have an easy tutorial-page for SVN? [16:20:47] <Xiong> may god bless and keep the czar -- far from here [16:25:55] <fundamental> Is there any reason why a link to a project within a wiki would remain a red link after the page has been created, edited, and even subpages created? [16:31:54] <fundamental> If anyone has an idea why it might be having the problem the red link can be seen here: http://slax.scribblewiki.com/Featured_Module [16:33:13] <Nikerabbit> fundamental: look at the link target!! [16:34:10] <fundamental> The target was [[Project:Approval]] which directs the user to the right page, but in edit mode. [16:34:39] <Nikerabbit> fundamental: how about setting more reasonable namespace with $wgMetanamespace? [16:34:58] <Nikerabbit> can't test very much, your styles make my firefox go crasy [16:35:53] <fundamental> I am not the wiki's admin and don't even have sysop rights, but I also thought the name of the wiki might be the source of the problem. [16:36:48] <Nikerabbit> I guess periods in namespaces are the cause [16:38:36] <fundamental> Well, if there is no other solution I will try to contact the admin/sysop for the wiki to change the namespace variable if it is posible (it is hosted by a wikifarm). [16:50:27] <wikibugs> 04(REOPENED) Bot flags automatically although previous version is unflagged - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14196 (10pbirken) [16:51:13] <wikibugs> 03(mod) Bot flags automatically although previous version is unflagged - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14196 15enhancement->normal; +comment (10pbirken) [16:57:01] <rigel> MW keeps logging me out when i edit a page [16:57:21] <rigel> im using 1.12.0 i think [16:59:00] <Nikerabbit> is changePassword.php broken? [16:59:14] <rigel> how would i tell [16:59:39] <Nikerabbit> rigel: that wasn't an answer, sorry [16:59:46] <Nikerabbit> rigel: check your cookies [17:00:09] <rigel> it's doing it on three separate computers [17:00:10] <wikibugs> 03(NEW) Some bots do not flag revisions automatically - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14985 15enhancement; normal; MediaWiki extensions: FlaggedRevs; (pbirken) [17:00:16] <rigel> so im pretty sure its server side [17:00:41] <rigel> "keep me logged in" helps, but i dont want to have to check that [17:01:55] <Nikerabbit> rigel: shared hosting? [17:02:15] <rigel> yep [17:02:28] <rigel> is that the issue? [17:02:44] <Nikerabbit> rigel: which one? [17:02:51] <rigel> webfaction [17:10:37] <wikibugs> 03(mod) Microsoft SQL Server/MSSQL support (tracking) - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=9767 (10GarrettS) [17:11:21] <Nikerabbit> rigel: mm it may be because the place where php stores session data is not same on all machines [17:16:10] <Nikerabbit> ^demon|lunch: weird indendation [17:20:49] <cfedde> apparently dumpBackup.php was not available in wikimedia 1.4rc1. Is there a way to get a similar dump of all the pages from a mediawiki installation that is that old? [17:23:59] <Simetrical> cfedde, upgrade it and take the dump. :) [17:24:11] <Simetrical> You could copy the database somewhere else, upgrade the temporary installation, and dump from there if you like. [17:24:26] <Simetrical> In case you want to keep using 1.4rc1 for some reason . . . [17:24:44] <cfedde> Simetrical: apparently that's going to be a multi-step process since there are several schema changes since then. [17:24:53] <cfedde> But maybe that's the best approach... [17:26:04] <Simetrical> cfedde, just run update.php. It will apply all the schema changes in order. [17:26:24] <Simetrical> Every new release tests to make sure that upgrading from 1.2 or so still works and gives the right DB structure. [17:27:39] <cfedde> hum. Ok. I must have made some other mistake when I tried the upgrade on my test server. [17:28:08] <cfedde> I'll go back and try again. [17:28:27] <^demon> Nikerabbit: I swear you stalk my commits. [17:30:04] <brion> every once in a while we discover an exciting bug in the updaters, but it usually works yes :) [17:30:09] <brion> we do have some regression tests for it now [17:31:08] <Nikerabbit> ^demon: indeed [17:31:12] <^demon> brion: Are you aware bugzilla is dead? [17:31:18] <^demon> Can't ping it. [17:31:27] <^demon> Back now. [17:31:28] <^demon> That was odd. [17:31:41] <^demon> Nikerabbit: Seems you have a comment prepared for each commit I make :-P [17:32:20] <CIA-55> 03rotem * r38242 10/trunk/phase3/languages/messages/MessagesHe.php: Localization update for he. [17:32:23] <Nikerabbit> is it possible that an account is in a state where it doesn't accept the set password? [17:33:05] <brion> ^demon: yep [17:33:06] <brion> 10:24am] RobH: bugzilla is going down for 5 minutes folks [17:33:06] <brion> [10:24am] RobH: dont be alarmed [17:34:49] <^demon> That would be why I didn't know. [17:34:50] <^demon> ;-) [17:39:32] <Duesentrieb> 10:24am?... oh, west coast now :) [17:40:40] <brion> west coast times are the best [17:40:47] <brion> they're inconvenient for everybody else ;) [17:41:03] <mark> so you decided to move there huh :P [17:41:14] <mark> I'm gonna get you to get up at 4 am some day. [17:41:39] <Nikerabbit> can someone tell me what is going on. mediawiki says incorrect password for one user, even after resetting the password, and doesn't send email even when i set it too directly to the database [17:43:16] <Nikerabbit> ah [17:43:22] <Nikerabbit> the user was cached by apc [17:43:41] <Nikerabbit> and changePassword.php doesn't purge the cache! [17:43:56] <Nikerabbit> does someone know how to fix it, or should I file a bug? [17:50:03] <brion> hmm [17:51:01] <brion> Nikerabbit: it calls saveSettings() which calls clearSharedCache() [17:51:04] <brion> this should be sufficient [17:51:23] <Nikerabbit> weird [17:51:34] <Nikerabbit> I had to manually delete the entry from cache [17:51:38] <brion> maybe double-check that the $wgMemc->delete() works properly for apc [17:52:26] <brion> Nikerabbit: is this trunk or 1.12? [17:52:36] <brion> 1.12 hits the raw database and doesn't touch the cache [17:52:39] <Nikerabbit> brion: trunk [17:52:44] <brion> hmm should be fine then [17:56:04] <Nikerabbit> weird indeed [17:56:59] <CIA-55> 03jhsoby * r38243 10/trunk/extensions/ (2 files in 2 dirs): Adding StalkerLog to Translate, change wording of some messages in StalkerLog [17:57:54] <Nikerabbit> doesn't hit clearSharedCache if it is already there [17:58:04] <Nikerabbit> bah, does [17:58:08] <Nikerabbit> but doesn't clear [17:59:43] <Nikerabbit> object(FakeMemCachedClient)#3 (0) { [17:59:43] <Nikerabbit> } [17:59:47] <Nikerabbit> that may explain something [18:00:01] <mib_htwmmhdv> is there a freeware version of dreamweaver? [18:02:47] <Simetrical> mib_htwmmhdv, sure, Notepad. [18:03:17] <proletaire> Hi [18:03:19] <proletaire> gfjdkj [18:03:40] <Nikerabbit> brion: ahaha [18:03:47] <proletaire> Hi [18:04:10] <Nikerabbit> brion: I'm using APC, and changePassword is command line script [18:04:24] <proletaire> I need help (again) to install FlaggedRevs on my mediawiki [18:04:47] <proletaire> Can someone help me? [18:04:52] <Simetrical> !ask | proletaire [18:04:52] --mwbot-- proletaire: Don't say "I have a question", or ask "Is anyone around?" or "Can anyone help?". Just ask the question, and someone will help you if they can. Also, please read < http://workaround.org/moin/GettingHelpOnIrc > for a good explanation of getting help on IRC. [18:05:00] <Simetrical> proletaire, say what problems you're having and you'll find out. [18:05:58] <brion> Nikerabbit: *aieee* [18:06:01] <proletaire> To install FlaggedRevs, they ask to run the files maintenance/update.php from a command line, but I can't get to run those files via MS-DOS FTP [18:06:17] <brion> Nikerabbit: that sounds like a good reason to never use apc's caching :) [18:06:32] <Nikerabbit> brion: indeed [18:06:46] <Nikerabbit> "Mostly for testing and debugging. Setting this enables APC for the CLI version of PHP. Normally you wouldn't want to create, populate and tear down the APC cache on every CLI request, but for various test scenarios it is handy to be able to enable APC for the CLI version of APC easily." [18:07:19] <Simetrical> Oh, yeah, APC has separate cache for CLI and web, doesn't it? Ugh. [18:07:30] <Simetrical> This has actually come up as a problem for me before, in fact. [18:07:39] <Simetrical> But I forgot about it and switched to XCache anyway, for other reasons. [18:07:44] <Nikerabbit> mm [18:08:02] <Nikerabbit> why can't it use one global cache or something [18:09:32] <Nikerabbit> anyway, what is the idea of duplicating full php.ini in /etc/php5/{apache,cgi,cli}/ ? [18:10:59] <wikibugs> 03(NEW) fencepost error in footnote numbering - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14986 normal; normal; MediaWiki: Page rendering; (presto) [18:18:23] <yannf> why does my squid says "The requested URL /wiki/index.php/Main_Page was not found on this server." this looks as valid: the url in the log is http://www.wikilivres.info/wiki/index.php/Main_Page [18:50:32] <wikibugs> 03(mod) Extension Makebot is obsolete; should be removed - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14156 (10maxsem.wiki) [18:51:23] <wikibugs> 03(mod) Extension Makebot is obsolete; should be removed - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14156 +need-review +patch (10maxsem.wiki) [18:53:59] <wikibugs> 03(mod) Extension Makebot is obsolete; should be removed - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14156 (10raimond.spekking) [18:54:07] <wikibugs> 03(mod) Extension Makebot is obsolete; should be removed - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14156 (10Bryan.TongMinh) [18:56:06] <CIA-55> 03nikerabbit * r38244 10/trunk/extensions/Translate/scripts/ (cli.inc createMessageIndex.php): * Improvements to script output [18:58:13] <bretm> thomas [19:16:25] <ialex> brion: ping [19:18:31] <Nikerabbit> ialex: dung [19:18:42] <ialex> hi Nikerabbit [19:19:01] <dungodung> Nikerabbit! [19:19:34] <Nikerabbit> dungodung: what's up? [19:19:43] <dungodung> ceiling [19:20:23] <brion> ialex: pong [19:20:32] <brion> hey Jhs [19:20:38] <brion> and Nikerabbit and dungodung [19:20:41] <brion> and all you crazy kids [19:20:46] <brion> *hhhhuuuugggsssss* [19:20:50] <Addshore> *huggles* [19:20:55] <dungodung> hiii brion! :D [19:20:56] <Addshore> :O [19:21:01] <dungodung> shorthaird! [19:21:07] <ialex> brion: i found the problem for 18 of 19 failing tests (my commit of yesterday) [19:21:21] *Nikerabbit hugs brion's feet [19:21:32] <brion> kinky [19:21:42] <brion> ialex: yay [19:22:14] <ialex> brion: but for the last one (Inclusion of !userCanEdit() content), imo, the test is wrong [19:22:57] <brion> ok, what happens there? [19:23:16] <ialex> brion: it shows an edit section for protected template (MediaWiki: message) [19:23:26] <brion> hmmmm [19:23:39] <brion> well a general issue is how do these permission checks deal with the parser cache? [19:28:17] <Jhs> hey brion. are you drunk? [19:28:19] <Jhs> :P [19:28:55] <Duesentrieb> actually... the parser could just leave markers for the edit links, and the skin would inject them if appropriate [19:28:58] <brion> alas no, Jhs [19:29:29] <Jhs> but you wish you were? [19:29:51] <brion> coding's always better with beer [19:29:58] <brion> or a nice rum & coke [19:30:02] *brion has rum at home but no coke :( [19:30:07] <brion> this will not do [19:30:14] <brion> but at the office, nothin' [19:30:39] <Jhs> :D [19:31:52] <brion> hmmm, what's the proper format for literal ipv6 addresses as hostname in an email address? [19:31:58] <CIA-55> 03mark * r38245 10/trunk/ubuntu/autoinstall/common.cfg: Always boot from the first disk [19:32:03] <flyingparchment> user@[2001::1] [19:32:07] <flyingparchment> (same as ipv4) [19:32:19] <brion> ok, so same as use in URLs, with the [] around it [19:32:28] <flyingparchment> yes [19:32:38] <flyingparchment> but in email that's needed for v4 too, user@[] [19:33:04] <ialex> brion: for the page, it is checked, but for the templates, that the problem [19:33:36] <wikibugs> 03(NEW) Double redirects are automatically fixed even when checkbox is unchecked - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=14987 major; normal; MediaWiki: Redirects; (le.korrigan) [19:34:20] <wikibugs> 04(REOPENED) $wgServer lacks brackets in IPv6 URLs - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14977 +comment (10brion) [19:34:28] <brion> flyingparchment: crezzy [19:34:41] <Lady_Aleena> How do TOCs get generated? [19:36:25] <brion> they are born in fire and pain [19:36:37] <brion> what specifically are you asking about them? :) [19:37:52] <ialex> brion: should the rights also be in parser cache? (restriction levels) [19:38:01] <ialex> *parser cache key [19:38:15] <Lady_Aleena> brion, just wondering if it is a script or something else. [19:39:26] <brion> well, the parser produces the toc based on the headings it finds [19:39:41] <Lady_Aleena> Ah, so it is server side not client side. [19:39:48] <brion> ialex: well maybe, but how do you cleanly list that? [19:40:06] <brion> Lady_Aleena: right [19:40:20] <brion> not much happens client-side, we try to be as fully-functional w/o JS as possible :) [19:40:39] <ialex> brion: list restriction levels for which user can edit page/template [19:40:43] <Lady_Aleena> I wish I could have the code for it, but I wouldn't know how to impliment it on my site. [19:40:57] <Lady_Aleena> Plus, it wouldn't be usable locally for me. [19:41:06] <CIA-55> 03jhsoby * r38246 10/trunk/extensions/StalkerLog/StalkerLog.i18n.php: Changing one message in StalkerLog per siebrand's request [19:43:06] <ialex> $list = array(); foreach( $wgRestrictionLevels as $right ) if( $user->isAllowed( $right == 'sysop' ? 'protect' : $right ) ) $list[] = $right; $rights = implode( ',' $rights ); [19:43:48] <ialex> * implode( ',' $list ) [19:43:55] <ialex> +, [19:45:48] <CIA-55> 03brion * r38247 10/trunk/phase3/ (5 files in 4 dirs): [19:45:48] <CIA-55> Revert r38221, 38238 -- "Add new parser function {{apiurl}}. Also, add new global $wgApiScript because manually constructing the api script's path seemed like a bad idea." [19:45:48] <CIA-55> I don't really like this in general; the API isn't meant for the UI and there should be little to no call to link to it from body content. [19:45:48] <CIA-55> Additionally, I believe we're trying to move all new parser functions to the convention of using the # prefix to avoid conflict with the template namespace. [19:48:49] <CIA-55> 03jhsoby * r38248 10/trunk/extensions/StalkerLog/StalkerLog.i18n.php: Remove trailing period per siebrand. XD [19:54:14] <foofoobar> hi [19:55:21] <foofoobar> I'm going crazy trying to use the #tag feature of mediawiki 1.12.0 [19:55:25] <CIA-55> 03brion * r38249 10/trunk/phase3/ (4 files in 3 dirs): [19:55:25] <CIA-55> Revert r38229: [19:55:25] <CIA-55> "* Add a nice fieldset for consistency [19:55:25] <CIA-55> * Switch result line and input form [19:55:25] <CIA-55> * Do not show result line "Below is a list of 0 files...". We have the message 'noimages' instead." [19:55:25] <CIA-55> Fieldset label duplicates inputbox label, which looks kind of odd to me. [19:55:27] <CIA-55> Moving lines around when set is empty changes the click hotspot for "Show new images since..." link, which is pretty annoying when following new uploads [19:55:57] <CIA-55> 03simetrical * r38250 10/trunk/phase3/maintenance/parserTests.inc: [19:55:57] <CIA-55> Fix fatal error trying to run parser tests: [19:55:57] <CIA-55> Notice: Undefined variable: prevResults in /var/www/trunk/phase3/maintenance/parserTests.inc on line 1098 [19:55:57] <CIA-55> Fatal error: Unsupported operand types in /var/www/trunk/phase3/maintenance/parserTests.inc on line 1098 [19:56:04] <cirwin> foofoobar: why? [19:56:06] <Duesentrieb> foofoobar: what are you trying to use it for, and how? [19:56:37] <foofoobar> cirwin: trying to use it with the google-maps extension. http://rafb.net/p/bBcett54.html [19:56:53] <cirwin> what's the problem? [19:56:58] <foofoobar> trying to use it inside a template and it does not recognize any of the parameters.. [19:57:18] <foofoobar> trying and googling for 7h now [19:57:54] <cirwin> does it work if you use tag and put a value for street in manually? [19:58:27] <foofoobar> i used it in an earlier version, i had the parser hacked up - but the parser changed too much in the last release and thought tag would be easier and faster [19:58:45] <foofoobar> sec, trying [19:59:02] <yannf> http://www.mediawiki.org/wiki/Squid_Reverse_Proxy [19:59:10] <yannf> please correct if necessary [19:59:56] <foofoobar> cirwin: does not seem to work either, doh could have tried that! the author suggests to use tag [20:00:04] <cirwin> ok [20:00:08] <foofoobar> let me try without tag [20:00:11] <cirwin> yeah [20:00:42] <brion> scap time [20:01:05] <foofoobar> it works without tag, does not with tag [20:01:20] <cirwin> hmm [20:01:40] <cirwin> I am surprised you have no "" around {{{street}}} [20:01:57] <Jack_Phoenix> I have a special page extension, Special:GiftManagerLogo, that acts kinda like Special:Upload, but instead it stores the files in $wgUploadDirectory/awards. The extension is not working properly - when I try to upload a file to the server, I get the error: "Internal error Could not copy file /tmp/phpTk1yAB to file ." however, Special:Upload works well and so does Special:UploadAvatar. this is a part of SocialProfile extension and the differen [20:01:57] <Jack_Phoenix> ce between Special:GiftManagerLogo and Special:UploadAvatar is very minimal. what might be the likeliest cause of this error? I've checked the file permissions for $wgUploadDirectory/avatars & $wgUploadDirectory/awards and both have 775 [20:02:19] <Jack_Phoenix> (yikes, that was a tad bit long explanation of the problem) [20:02:25] <foofoobar> cirwin: http://rafb.net/p/qTPN6p78.html [20:02:40] <Sundance_Raphael> hi [20:02:57] <cirwin> foofoobar: ah! [20:03:04] <foofoobar> cirwin: i've tried about 100 different combinations - its supposed to be "{{{street}}}, {{{zip}}}" [20:03:07] <cirwin> you need a pipe on the start of each line [20:03:17] <cirwin> because you are sending named parameters [20:03:24] <cirwin> I thought you were sending one blob of configu [20:04:07] <foofoobar> is it because i'm not using {{{1}}}? [20:04:11] <cirwin> try http://rafb.net/p/2EFzfP18.html [20:04:16] <foofoobar> because comment 72 in this https://bugzilla.wikimedia.org/show_bug.cgi?id=2257 hasnt got them [20:04:35] <cirwin> yeah [20:04:43] <cirwin> inputbox takes config from plaintext [20:04:49] <cirwin> not from parameters [20:05:09] <cirwin> actually, you should remove the "s from the example I pasted [20:05:59] <foofoobar> oh man! you're the best! thank you very much! [20:06:10] <cirwin> np [20:06:15] <foofoobar> i wouldnt have found this bug, i had the pipes - but everything on one line [20:06:24] <cirwin> ok [20:06:27] <foofoobar> thank you very much :) [20:06:31] <foofoobar> was about to pull my hair out [20:06:33] <cirwin> any time [20:06:36] <ialex> brion: any idea on what I proposed before? [20:07:05] <cirwin> Jack_Phoenix: it's kind of hard to help with an extension that I've never used - if you say the permissions are set right, I don't have any other ideas [20:07:19] <foofoobar> just checked, the help page of the extension has it wrong too [20:07:31] <cirwin> foofoobar: {{sofixit}} :P [20:08:01] <Jack_Phoenix> yeah, that's making me go "wtf" too since Special:Upload etc. work properly on this wiki [20:10:25] <foofoobar> done :) [20:11:10] <cirwin> yay [20:11:26] <cirwin> now hopefully we won't have someone here tomorrow asking the same thing ;) [20:15:07] <CIA-55> 03brion * r38251 10/trunk/phase3/ (RELEASE-NOTES includes/DefaultSettings.php): [20:15:07] <CIA-55> Revert r38214 (IPv6 check for $wgServerName) -- failing to load IP class in some situations as seen on Wikimedia live installation. [20:15:07] <CIA-55> Haven't been able to reproduce locally, though. [20:16:28] <MZMcBride> cirwin: When people ask the same thing every day, it makes me look smarter. ;) [20:16:46] <wikibugs> 03(mod) Some bots do not flag revisions automatically - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14985 (10JSchulz_4587) [20:17:14] <brion> ialex: looks kind of icky to me imho [20:17:28] <wikibugs> 03(mod) $wgServer lacks brackets in IPv6 URLs - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14977 (10brion) [20:18:24] <AaronSchulz> brion: I love how bugs always pile up after long periods of silence ;) [20:18:39] <ialex> brion: what would you propose? [20:20:22] <brion> can we get php & apache to log URLs with their damn fatal/segfault errors? [20:21:30] <brion> ialex: i dunno, but just checking some protection levels will be insufficient if there are additional per-namespace or per-page restrictions on editing [20:24:04] <wikibugs> 03(FIXED) Set $wgBlockAllowsUTEdit for Greek Wikipedia - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=10304 +comment (10brion) [20:27:14] <wikibugs> 03(mod) Use Preprocessor_Hash by default to avoid missing DOM module errors - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13770 (10brion) [20:32:58] <ialex> brion: per-page restrictions should be catched by my check [20:34:00] <yannf> now that squid is working, how to setup it so that requesting twice the Main_Page gives a TCP_HIT instead of a TCP_MISS ? [20:35:00] <ialex> per namespace too with the default config [20:37:10] *dungodung stumps brion about global blocking and hides :D [20:37:33] <brion> if it's not about segfaulting apaches i don't care right now [20:37:47] <DuTempete> Is anyone familiar with the StringFunctions extension? [20:37:49] <dungodung> tis not, unfortunately [20:38:10] <DuTempete> I'm getting this error: Warning: Cannot modify header information - headers already sent by (output started at E:\www\wiki\extensions\StringFunctions\StringFunctions.php:223) in E:\www\wiki\includes\WebResponse.php on line 10 [20:41:24] <cirwin> DuTempete: have you looked at line 223 of the file? [20:41:59] <DuTempete> not sure I can determine which line that is, but I'll try if you think it'll help [20:42:28] <cirwin> hmm [20:42:42] <cirwin> it is just a function definition [20:42:49] <cirwin> did you edit the file at all? [20:43:15] <DuTempete> No, I haven't touched it [20:43:32] <cirwin> wierd [20:44:26] <DuTempete> ooh, sec, lemme test something [20:44:30] <CIA-55> 03aaron * r38252 10/trunk/extensions/FlaggedRevs/FlaggedRevision.php: Remove call to bogus var [20:44:54] <DuTempete> the line before the one that calls StringFunctions, in the localsettings, effects headers [20:44:56] <wikibugs> 03(mod) Some bots do not flag revisions automatically - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14985 (10pbirken) [20:46:10] <DuTempete> nevermind... removed it and still getting the error [20:47:53] <yannf> can anyone help optimise a squid setup please? [20:47:56] <DuTempete> As far as I can tell, this is line 223: <table cellpadding="2" cellspacing="0" class="auto"> [20:48:27] <CIA-55> 03rotem * r38253 10/trunk/phase3/languages/messages/MessagesHe.php: Localization update for he. [20:54:17] <srastin> When I use Google Mini/Google Search Appliance to index my MediaWiki site it drives CPU usage to the upper 90% and it stays up there. The problem appears to be related to Google reindexing all pages every time it hits them. It doesn't realize they have not changed. Any ideas/suggestions how I can bring this problem under control, other than throttling back the Google box? [21:02:41] <CIA-55> 03simetrical * r38254 10/trunk/phase3/includes/ (Linker.php Sanitizer.php): New Linker::link() function, intended to replace Linker::make*Link*() functions. Cleaner interface and more cleanly written. Convert a bunch of stuff to use it. All parser tests pass (except the usual 17). [21:04:14] <DuTempete> cirwin: I had installed it incorrectly [21:04:22] <cirwin> ok [21:04:27] <cirwin> all working now? [21:04:40] <Simetrical> srastin, you can try using a sitemap, I guess? [21:04:48] <Simetrical> Google has to look at the page to see if it changed, generally. [21:04:55] <Nikerabbit> Simetrical: and :D [21:04:58] <DuTempete> I didn't realise I was supposed to open the file and copy the contents, rather than saving the whole file directly from the trunk [21:06:20] <cirwin> ok [21:07:11] <wikibugs> 03(mod) Use Preprocessor_Hash by default to avoid missing DOM module errors - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13770 (10Simetrical+wikibugs) [21:11:30] <CIA-55> 03simetrical * r38255 10/trunk/phase3/includes/Linker.php: Add a comment, and fix up another. Writing the documentation before you write the code is nice, until you scale back what the code tries to achieve and the documentation becomes wrong. :) [21:16:38] <Lady_Aleena> Could someone please look at the following parser function and tell me if I have closed it properly? I think I did, but I may not have. [21:16:48] <Lady_Aleena> {{#ifexist:Category:{{{PROJECT}}} infoboxes|{{{PROJECT}}} infoboxes|{{#ifexist:Category:{{{PROJECT}}} templates|{{{PROJECT}}} templates|Infobox templates}}}} [21:17:30] <Duesentrieb> seems like it [21:17:51] <Lady_Aleena> Well, that is not the problem then. [21:18:15] <Lady_Aleena> I really wish I felt comfy enough to swear in here. [21:19:09] <Duesentrieb> sometimes it *would* be helpfull to generate syntax errors... [21:19:15] <Lady_Aleena> I just want to know why this blasted table isn't closing. [21:19:21] <Duesentrieb> "everything is valid" comes back to hount you [21:19:30] <Duesentrieb> Lady_Aleena: link? [21:19:48] <Lady_Aleena> http://en.wikipedia.org/wiki/User:Lady_Aleena/Test_banner is the template [21:20:00] <Lady_Aleena> http://en.wikipedia.org/wiki/User_talk:Lady_Aleena/Test_banner shows that it is broken [21:20:25] <wikibugs> 14(INVALID) Patch to make Special: Statistics result rendering consistent with Special:Search - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14983 +comment (10brion) [21:20:40] <Duesentrieb> it does? [21:21:06] <Duesentrieb> broken how? [21:22:18] <Lady_Aleena> The table isn't closing. [21:22:34] <Lady_Aleena> I have |} at the end, but when transcluded, the table does not close. [21:23:54] <Lady_Aleena> On the talk page, there are supposed to be two (2) different boxes, 1 as the template alone and the other as the template within a template. [21:23:57] <CIA-55> 03simetrical * r38256 10/trunk/extensions/Cite/Cite_body.php: Delete Cite::validateName(). Use Xml::escapeId() instead. All parser tests pass. [21:24:11] <Simetrical> brion, could I please delete the 17 tests that have never actually passed? Because it really makes it annoying to figure out if you've actually broken anything. [21:24:23] <brion> Simetrical: NO [21:24:27] <Simetrical> :'( [21:24:31] <brion> just run parserTests.php --record [21:24:34] <brion> easy as pie [21:24:44] <brion> tells you exactly what changed [21:24:54] <Duesentrieb> Lady_Aleena: well... you basically have to check all {{...}} pairs on that page :) [21:25:11] <Lady_Aleena> I have over and over and over again. [21:25:19] <Simetrical> It still lists all the old failing tests, with diffs if I want to enable diffs. [21:25:21] <Lady_Aleena> I have torn it apart and put it back together. [21:25:36] <Simetrical> I have to hunt through 17 useless diffs to get to the ones I actually want. [21:26:00] <Duesentrieb> Lady_Aleena: but it's not right. my editor tells me that the opening { for the last } ist at {{#if instead of at {| [21:26:05] <Duesentrieb> so... they don't match [21:26:07] <Simetrical> And it fills up my whole terminal screen even without diffs. [21:26:18] <brion> Simetrical: look at the end [21:26:24] <brion> where it tells you exactly what changed [21:26:26] <brion> then just look at thsoe [21:26:29] *Simetrical grumbles [21:26:31] <Lady_Aleena> Duesentrieb, can you tell at which #if? [21:26:54] <Duesentrieb> the last one. but it's only a symptom [21:26:58] <Simetrical> Diffs don't even seem to work with --record. [21:27:00] <Duesentrieb> the cause appeasr to be this: {{{!}} [21:27:07] <Duesentrieb> there's a { too much there... [21:27:12] <blobaugh> ive been searching, is there a way to import the info into mediawiki from tikiwiki? [21:27:14] <Simetrical> --compare --record --quiet shows no diffs. [21:27:30] <Duesentrieb> removing that seems to match up all. [21:27:36] <Duesentrieb> Lady_Aleena: i'll try to fix it [21:28:02] <Lady_Aleena> Duesentrieb, thanks!!! [21:28:16] <Duesentrieb> Lady_Aleena: something else is still wrong, but it's matched up at least now :) [21:28:34] <Duesentrieb> Lady_Aleena: a decent editor with stuff like parantecy matching does help sometimes :) [21:29:05] <Lady_Aleena> Duesentrieb, I am using a trial version of EditPad, so I can't add anything unless I pay for it. [21:29:23] <Duesentrieb> use something free, then :) [21:29:37] <Duesentrieb> or at least, freeware with decent features [21:29:45] <Duesentrieb> parantacy matching is a min [21:30:27] <Lady_Aleena> Duesentrieb, I can't find anything else like it. [21:30:56] <Duesentrieb> EditPlus&UltraEdit are shareware [21:31:03] <Duesentrieb> SKiTE and JEdit are Free [21:31:26] <Duesentrieb> Never looked at Notepad++, i think it's freeware too [21:31:29] <wikibugs> 14(WFM) Use localized alias of Special:Search consistently - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14972 +comment (10brion) [21:32:09] <Lady_Aleena> If only uninstalling software were really easy. [21:32:14] <Duesentrieb> it's SciTE, sorry [21:32:22] <Duesentrieb> well, on linus, it is :) [21:32:25] <Duesentrieb> linux [21:32:28] <Duesentrieb> damn, need to sleep [21:32:50] <Duesentrieb> apt-get remove crap; apt-get install goodstuff [21:32:52] <Duesentrieb> done :) [21:34:06] <Lady_Aleena> Duesentrieb, I use Window$ XP [21:34:31] <Duesentrieb> well, in theory, you go to add/remove software, and click a few buttons :) [21:34:56] <wikibugs> 14(WFM) Resize to thumbnail is ugly for PNG with transparency/ alpha channel - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=6205 +comment (10brion) [21:36:26] <Lady_Aleena> Duesentrieb, in theory yes. [21:36:40] <Lady_Aleena> That doesn't mean that stuff isn't left behind. [21:37:15] <Duesentrieb> sure. so just leave it and ignore it. and get something better :) [21:37:31] <CIA-55> 03nad * r38257 10/trunk/extensions/CSS/CSS.php: should use htmlspecialchars here [21:37:42] <Lady_Aleena> Duesentrieb, I need to save disk space. [21:37:54] <CIA-55> 03yaron * r38258 10/trunk/extensions/SemanticGoogleMaps/SGM_QueryPrinter.php: Fixed error in setting zoom - longitude goes from -180 to 180, not -90 to 90 [21:38:07] <CIA-55> 03yaron * r38259 10/trunk/extensions/SemanticGoogleMaps/SemanticGoogleMaps.php: New version, 0.3.2 [21:38:30] <Klemo> Hello wiki fans... [21:38:52] <Duesentrieb> Lady_Aleena: editplus is about 1mb :) [21:39:07] <Klemo> I can't seem to get the list-style-type square to show in my media wiki template for toclevel-1 using simple as my starting base [21:39:48] <Duesentrieb> Klemo: try providing an *absolute* path for the image [21:39:59] <Lady_Aleena> Duesentrieb, do you use it? [21:40:05] <sherrod> hey guys, I am wondering if there is a channel or a community where I can find Mediawiki developers who do paid side work on MW. Is there anything like that? [21:40:13] <Duesentrieb> Lady_Aleena: i used to, back in the day when i was using windows [21:40:24] <Duesentrieb> like, five years ago :) [21:40:46] <Duesentrieb> sherrod: not that i know of [21:40:48] <wikibugs> 03(mod) AntiSpoof should return more than one result - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=12232 (10brion) [21:41:14] <Duesentrieb> sherrod: try writing to the mailing list. [21:41:31] <sherrod> Duesentrieb: where is the mailing list? [21:41:41] <Duesentrieb> !mail [21:41:41] --mwbot-- I don't know anything about "mail". You might try: !email !enotif !mailinglists !parser !smtp !smw !support !worstcase [21:41:49] <Duesentrieb> !mailinglists [21:41:49] --mwbot-- See http://lists.wikimedia.org/ [21:41:57] <Duesentrieb> the mediawiki-l one [21:42:03] <Duesentrieb> you have to subscribe to post though [21:42:14] <sherrod> ok, cool, I will try that [21:42:24] <Duesentrieb> you should be on mediawiki-announce anywayif you are running mw [21:42:30] <Duesentrieb> it tells you about updates and stuff [21:43:29] <CIA-55> 03brion * r38260 10/trunk/extensions/CSS/CSS.php: Revert r38257 "should use htmlspecialchars here" -- this *removes* a call to htmlspecialchars(), the opposite of what is described. WTF? [21:48:36] <brion> Simetrical: linkUrl() seems like it's duplicating wfArrayToCgi() [21:49:10] <Simetrical> brion, I had forgotten about that. Does it seem good otherwise? [21:49:27] *OverlordQ goes to look at 12232 again [21:49:29] <brion> also the if( isExternal()) getFullUrl() else getLocalUrl() should be unnecessary [21:49:33] <brion> getLocalUrl() already does this [21:49:36] <Simetrical> Yeah, I just noticed that. [21:49:37] <brion> or should :) [21:50:39] <brion> as a general whinge about coding standards, we use $foo[] = blah rather than $foo []= blah [21:51:03] <Klemo> not an image I want to use squre... [21:51:12] <Klemo> list-style-type: square; [21:51:24] <Simetrical> But []= is logically an assignment operator. :( [21:51:25] <Klemo> I cant figure out how to assign it to toclevel-1 [21:51:36] <Duesentrieb> ah, i though you meant the scare bullet used by monobook [21:51:40] <Duesentrieb> which is indeed an image [21:51:47] <Duesentrieb> err, *square [21:52:06] <brion> Simetrical: Sanitizer::mergeAttributes() has the special handling for the 'class' attrib [21:52:13] <Klemo> .toclevel-1 { list-style-type: square; } [21:52:14] <brion> you might want to use that rather than duping it [21:52:18] <Klemo> does not work [21:52:19] <brion> in linkAttribs [21:52:22] <Duesentrieb> Klemo: .toclevel-1 { list-style-type: square; } doesn't do it? does it need to be assigned to the list or the item? [21:52:28] <CIA-55> 03simetrical * r38261 10/trunk/phase3/includes/ (Linker.php Title.php): As Brion points out, Linker::linkUrl() duplicates wfArrayToCGI. Fix that, and also respace/expand a comment in Title.php. [21:52:31] <Simetrical> brion, better and better. [21:52:33] <Lady_Aleena> OMG! I did it! Oh thank you Duesentrieb...you showed me where there was an "error" but it wasn't an error as we thought, it was the opening of another table within. All I had to do was close that table, and it works like I wanted! Thank you thank you thank you! [21:52:33] <Klemo> a line item [21:52:42] *Simetrical waits for the critique to finish before committing again [21:52:44] <Klemo> tag is <li class=toclevel-1... > [21:53:14] <Duesentrieb> Lady_Aleena: ah, i see - you replaced one too many | by {{!}}, resulting in {{{!}} instead of {| :) [21:53:28] <Klemo> <li class="toclevel-1"> [21:53:55] <Mr_Grieves> Hi, I wonder how I may create a "featured article of today"-template (No voting, just randomly selecting one). [21:53:58] <Lady_Aleena> Duesentrieb, yes, and all I needed was at the end {{!}}} :) [21:54:15] <Klemo> wow busy day on the mediawiki irc [21:55:12] <Mr_Grieves> I've tried to google it several days.. but google's indexing of 1000000 wiki's does help :-) [21:55:41] <Mr_Grieves> I was hoping there was a neat place where you could download common templates.. [21:55:48] <Duesentrieb> Mr_Grieves: what you see on wikipedia is no magic. it's simply done by hand. [21:55:51] <wikibugs> 03(mod) Double redirects are automatically fixed even when checkbox is unchecked - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14987 (10sb_paramount) [21:56:11] <Klemo> Duesentrieb: .toclevel-1 { list-style-type: square; } does not work, I still get plain list of items [21:56:20] <Duesentrieb> Mr_Grieves: you can tryto steal nice templates from other wikis, but wikipedia, especially en.wikipsia, is notorious foir complex and messy templates [21:56:34] <Duesentrieb> having a nice set of standard templates ready for download is something i have been thinking aboitu for a while [21:56:38] <Duesentrieb> but it doesn't exist afaik [21:56:40] <Klemo> I've started with the simple skin, and started modifiing from there, perhaps somehing is display: none before hand? [21:56:47] <yannf> can anyone help optimise a squid setup please? [21:56:55] <Duesentrieb> Klemo: get firebug and look. [21:57:01] <Klemo> will do [21:57:03] <skierpage> Firefox 3.1 nightly now directly supports OGG Theora and Vorbis in <audio> and <video> tags! The end of Java and/or plugin players is in sight. [21:57:19] <Duesentrieb> Klemo: firebug is an absolute lifesaver when debugging css [21:57:24] <yannf> http://www.mediawiki.org/wiki/Squid_Reverse_Proxy i get too many TCP_MISS [21:57:34] <skierpage> Does anyone here have bleeding edge Safari 3.1? If so, does anything on http://www.skierpage.com/moz_bugs/test_audio_video_tags.html work for you? [21:57:38] <Duesentrieb> skierpage: not for a loooooong time it ain't [21:57:59] <Klemo> yeah I know what it is... I just for get about it, on the another hand, if you enable the Developer menu in Safari it does some cool stuff tooo [21:57:59] <Duesentrieb> we need to be cave-compatible [21:58:00] <skierpage> Duesentrieb, I have a dream [21:58:06] <skierpage> :-) [21:58:30] <Lady_Aleena> Duesentrieb, what text editor do you use again? [21:58:38] <Duesentrieb> i think there's still compat code for ie 5.1/m [21:58:49] <Duesentrieb> err, 5.1/mac in mediawiki [21:58:49] <Klemo> crap... fire bug requires firefox 3... boo [21:59:01] <Simetrical> Klemo, uh, get an old version? [21:59:02] <Duesentrieb> Klemo: try to get an old version [21:59:17] <Duesentrieb> ff3 is pretty unstable for me [21:59:19] <skierpage> http://getfirebug.com/releases/ says FF2 also [21:59:21] <Klemo> I'll take a lookin safari [21:59:30] <Duesentrieb> rarely crashesm but frequently freezes... [21:59:30] <Mr_Grieves> Duesentrieb, ah, I was hoping not having to learn about template for "real". [21:59:56] <Duesentrieb> Mr_Grieves: you can try to just copy, but it'll bite you at some point [22:00:30] <Duesentrieb> but yea... maintaining a "bag of templates" on mediawiki.org would be nice... [22:01:42] <Mr_Grieves> Duesentrieb, I'm not even sure how to copy a template.. ehe. Is http://www.mediawiki.org/wiki/Template the ultimate "create you're own template resource"? [22:02:39] <CIA-55> 03simetrical * r38262 10/trunk/phase3/includes/ (Linker.php Sanitizer.php): [22:02:39] <CIA-55> Use Sanitizer::mergeAttributes() for Linker::linkAttribs(). Also clean up [22:02:39] <CIA-55> whitespace for mergeAttributes and reduce number of nested functions, and don't [22:02:39] <CIA-55> try to merge non-string 'class' arguments. This last point is necessary so I [22:02:39] <CIA-55> can have 'class' => false work right for linkAttribs(), but it makes sense. [22:02:41] <CIA-55> Parser tests pass. [22:03:21] <Mr_Grieves> I tried to find variables useful for creating my own "featured article"-template, but at a glace I couldn't really see what would fit.. [22:03:57] <Mr_Grieves> Just one question, is it possible to create such a template with Wiki mark-up language? [22:04:40] <Mr_Grieves> Or would it be simpler to write something in .php (which I do know).. [22:05:44] <brion> Simetrical: "Parser tests pass" is music to my ears :D [22:06:41] <blobaugh> is there any way to import from tikiwiki? [22:06:49] <Duesentrieb> !templates | Mr_Grieves [22:06:49] --mwbot-- Mr_Grieves: For more information about templates, see <http://www.mediawiki.org/wiki/Help:Templates>. The most common issues with templates copied from Wikipedia can be fixed by installing ParserFunctions <http://meta.wikimedia.org/wiki/Help:ParserFunctions> and enabling HTML Tidy <http://www.mediawiki.org/wiki/Manual:%24wgUseTidy>. [22:06:54] <Duesentrieb> !export | Mr_Grieves [22:06:54] --mwbot-- Mr_Grieves: To export pages from a wiki, navigate to Special:Export on the wiki, type in the names of the pages to export, and hit "export". See <http://en.wikipedia.org/wiki/Special:Export> for an example of this form. See also: !import [22:06:58] <Duesentrieb> !import | Mr_Grieves [22:06:58] --mwbot-- Mr_Grieves: To import pages a few pages, use Special:Import - you can also import pages directly from another wiki (see !importsources). For mass imports, use importDump.php - see <http://www.mediawiki.org/wiki/Manual:Importing_XML_dumps> for details. NOTE: when using content from another wiki, follow the LICENSE TERMS, especially, attribute source and authors! [22:07:34] <Mr_Grieves> Duesentrieb, That was very helpful, thank you so much. [22:08:30] <Duesentrieb> blobaugh: http://www.google.com/search?hl=en&q=tikiwiki-to-mediawiki&btnG=Suche&meta= [22:10:04] <DanielTahar> how do I create a "calculator" which calculates someone's age? i started a brand new wiki so i have nothing from wikipedia [22:10:09] <Klemo> thanks [22:10:24] <Klemo> I was inheriting a list-type: none from above... [22:10:31] <Klemo> good by now [22:11:20] <cirwin> DanielTahar: see http://meta.wikimedia.org/wiki/Help:ParserFunctions [22:11:47] <Duesentrieb> DanielTahar: wikis are for managing text, not for calculating stuff [22:12:02] <DanielTahar> yeah, you know what i mean, it's not really "calculating"... [22:12:05] <Duesentrieb> cirwin: i'm doubtfull that parserfunctions will do what he wants :) [22:12:12] <cirwin> it does on 'pedia [22:12:19] <Duesentrieb> well, maybe, then they are what you want. [22:12:41] <blobaugh> Duesentrieb: that doesnt help me. i havent seen anything that shows an upgrade [22:12:44] <Duesentrieb> cirwin: well... yes. ased on page content. but not interactively - which is what i would call a "calculator". [22:13:04] <cirwin> Duesentrieb: we'll find out, no doubt :) [22:13:36] <Duesentrieb> blobaugh: no idea, sorry. i mograted an erfurtwiki some time ago, wrote myself a bit of perl to convert the most basic syntax. worked nicely [22:13:51] <Duesentrieb> i expect the same can be done for tiki. but manual fixing will be needed [22:13:56] <Duesentrieb> oh yay. [22:15:28] <Klemo> Ok I found the problem... [22:15:29] <blobaugh> Duesentrieb: bummer. i dont really want to manually make some magic. only gonna be at this job for a few more days so it is not really worth my time [22:16:11] <Duesentrieb> keep digging... [22:16:12] <Klemo> when I had multipule sections == 1 == === 1.1 === .... 1.1 would show the square, and section 1 would show nothing [22:16:31] <Klemo> it was showing the squre for section 1 it was just off the left of the page [22:16:39] <Klemo> I added some padding and its there now... [22:16:57] <blobaugh> Duesentrieb: although it would be a good tool to have and post on my website for others.... [22:16:59] <Simetrical> brion, so those last two failing tests are wrong, I think. The first two are even more wrong: localurl has never URL-encoded parameters, and it would break everything if it did. [22:17:10] <brion> bleh :) [22:17:12] <Klemo> thanks again [22:17:15] <Simetrical> brion, there are also at least two failing tests where the *desired output* is invalid XML. [22:17:25] <brion> heh that might be bad ;) [22:17:38] <Duesentrieb> blobaugh: maybe you find something here: http://meta.wikimedia.org/wiki/Special:Search/Tikiwiki [22:17:39] <Klemo> biron! [22:17:40] <brion> probably at least some of them are wrong yes [22:17:43] <brion> kelmo! [22:17:46] <Duesentrieb> blobaugh: much of that will be stale though [22:17:48] <Klemo> lol [22:17:51] <Klemo> bye now [22:17:55] <ryno1> I am integrating MediaWiki into Drupal and I am getting the following error when I run it locally. Any ideas or experience on why this happens? [22:17:55] <ryno1> Fatal error: Call to a member function getConnection() on a non-object in C:\wamp\www\AtTask\sites\all\modules\mediawiki\application\includes\GlobalFunctions.php on line 2304 [22:17:55] <ryno1> Thanks in advance! [22:18:23] <Simetrical> -<td><a href="ftp://|x||" class="external autonumber" title="ftp://|x||" rel="nofollow">[1]</td><td></a>" onmouseover="alert(document.cookie)">test [22:18:30] <Simetrical> <td></a>? [22:18:36] <Simetrical> brion, can I remove the obviously wrong ones? [22:18:56] <brion> Simetrical: hmm, better to correct them probably [22:18:58] <blobaugh> Duesentrieb: already looked there.. [22:18:59] <wikibugs> 14(DUP) Allow user_talk edits for blocked users in el.wikipedia - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14603 +comment (10geraki) [22:19:01] <Simetrical> brion, okay. [22:19:10] <brion> now that case there was one where there was a js injection, which was fixed [22:19:18] <brion> but the other parts of the parsing probably were still wrong :) [22:19:40] <brion> so fix that up to current behavior i guess [22:19:43] <blobaugh> Duesentrieb: i really should write an import script though. it would be useful to many cause i have seen the same issue in many places [22:19:47] <brion> and double-check there's no regression on the js ;) [22:19:49] <wikibugs> 03(mod) Set $wgBlockAllowsUTEdit for Greek Wikipedia - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=10304 (10geraki) [22:19:54] <Duesentrieb> blobaugh: http://brightbyte.de/repos/codebin/ew2mw.php?view=1 is what i used (php, no perl after all). syntax differes, but the principle is the same [22:20:00] <OverlordQ> oh wow I'm an idiot. I spend five minutes wondering why the option for wfMsgExt was parsemag, should it be parsemsg? Till i realized the mag stood for magic. [22:20:46] <ryno1> Anyone have experience with Drupal/MediaWiki? [22:21:10] <blobaugh> Duesentrieb: thanks [22:22:34] <wikibugs> 03(NEW) Request for activating the WikiTimeLine extension on wikipedia or sister project - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14988 15enhancement; normal; MediaWiki extensions: Extensions requests; (echalone) [22:26:01] <wikibugs> 14(INVALID) Request for activating the WikiTimeLine extension on wikipedia or sister project - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14988 +comment (10conrad.irwin) [22:26:27] *cirwin has never closed a bug before so someone important might want to double check [22:27:26] <wikibugs> 03(mod) mwsuggest.js: Case-insensitive search - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14399 (10brion) [22:27:41] <Lady_Aleena> Does Mozilla have a text editor? [22:29:07] <wikibugs> 03(mod) Request for activating the WikiTimeLine extension on wikipedia or sister project - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14988 (10echalone) [22:31:17] <wikibugs> 03(mod) Add rd_interwiki colum to redirect table - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14418 (10brion) [22:39:54] <CIA-55> 03simetrical * r38263 10/trunk/phase3/maintenance/parserTests.txt: [22:39:54] <CIA-55> Change the results for 6 out of 17 failing parser tests so that they pass. Now only 11 tests fail for me. The remaining outcomes are: [22:39:54] <CIA-55> * Invalid XML (this is nine of them) [22:39:54] <CIA-55> * Double apostrophes in a link [22:39:54] <CIA-55> * Something about language variants that I don't understand and so will leave alone [22:51:30] <Lady_Aleena> Gotta go for a bit... [22:59:46] <wikibugs> 03(NEW) batchAntiSpoof.php fails with Postgres - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14989 minor; normal; MediaWiki extensions: AntiSpoof; (overlordq) [23:12:10] <Jhs> brion, would you mind if added accesskey="s" to the submit button of various forms? [23:12:46] <Jhs> i think you commented on a bug of mine that you didn't want to do it (for renameuser i think), because it would be very incomplete [23:13:47] <wikibugs> 03(mod) AntiSpoof should return more than one result - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=12232 +comment (10overlordq) [23:14:07] <wikibugs> 03(mod) Enable "What links here" on applicable special pages - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14787 (10brion) [23:14:11] <brion> Jhs: if it doesn't conflict with things... [23:14:46] <Jhs> of course [23:14:55] <wikibugs> 03(mod) AntiSpoof should return more than one result - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=12232 +comment (10overlordq) [23:15:51] <OverlordQ> Alright that's enough attempts at that bug lol I give up after this one. [23:19:51] <wikibugs> 03(mod) Special:Export remove "Category:" (or local name) from text box - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14790 15enhancement->minor; +comment (10brion) [23:20:06] <brion> anybody want to try 14790? it's probably a real easy fix in SpecialExport.php [23:20:27] <exobuzz> brion: hows things since rc1 ? many problems or mostly ok ? [23:20:31] <brion> i added a detailed explanation of the problem, which wasn't initial clear ot me [23:20:42] <brion> exobuzz: pretty much so far so good i think :) [23:20:49] <exobuzz> brion: nice one. ;-) [23:20:52] <brion> might still be an issue with the "crappy RHEL installation" things [23:22:17] <exobuzz> brion: so on 14790 the code basically just needs to strip Category: if the prefix is there. [23:22:46] <wikibugs> 03(mod) Add RSS with all items - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14794 (10brion) [23:22:47] *exobuzz takes a look [23:22:57] <brion> exobuzz: well it probably just needs a Title::newFromText( $blah, NS_CATEGORY ) [23:25:46] <wikibugs> 03(mod) Automatic hyphens to (localized?) dashes - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=1485 +comment (10brion) [23:26:30] <exobuzz> it has a maketitlesafe.. that needs to go first ? [23:27:21] <brion> lemme look [23:27:23] *exobuzz needs to read the api [23:27:26] <wikibugs> 03(mod) $wgServer lacks brackets in IPv6 URLs - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14977 (10innocentkiller) [23:27:49] <brion> exobuzz: i think i'd just replace the makeTitleSafe with a newFromText [23:28:02] <exobuzz> yeh.. ok.. let me post a diff for this then [23:28:03] <exobuzz> ;-) [23:28:08] <brion> (order of params will be different -- $catname is the text, NS_CATEGORY is the default namespace) [23:28:15] <brion> but i think it'll do the job [23:28:23] <exobuzz> ill just test it locally and then post a diff [23:28:28] *brion hmms [23:28:46] <exobuzz> or you can just fix it ;-) [23:28:54] <exobuzz> i just felt like doing something :) [23:29:00] <brion> though that might not work right for things like Category:Wikipedia:Translation [23:29:07] <brion> bother :) [23:29:20] <exobuzz> what would that call return in that vase ? [23:29:32] <exobuzz> case [23:29:36] <ryno1> I am getting the following error : [23:29:37] <ryno1> Fatal error: Call to a member function getConnection() on a non-object in C:\wamp\www\AtTask\sites\all\modules\mediawiki\application\includes\GlobalFunctions.php on line 2305 [23:29:37] <ryno1> What does that even mean? [23:29:41] <brion> i _think_ the Wikipedia: would get recognized as project namespace and override the default Category: [23:29:52] <brion> ryno1: it means something failed [23:29:58] <brion> look on line 2305 of GlobalFunctions.php [23:30:06] <brion> whatever it was that it's calling on isn't an object [23:30:18] <brion> probably a failure to set up the load balancer, possibly due to a bad configuration or a bad db [23:30:22] <ryno1> It is the [23:30:22] <ryno1> function &wfGetDB( $db = DB_LAST, $groups = array() ) { [23:30:22] <ryno1> global $wgLoadBalancer; [23:30:22] <ryno1> $ret = $wgLoadBalancer->getConnection( $db, true, $groups ); [23:30:22] <ryno1> return $ret; [23:30:23] <ryno1> } [23:30:31] <ryno1> yeah that is exactly what it is [23:30:46] <ryno1> What could the possible problems be? [23:31:53] <exobuzz> where is the mediawiki code api/tree thing [23:32:00] <exobuzz> showing all the calls etc [23:32:07] <ryno1> when I do a var_dump on global $wgLoadBalancer; it returns null...is that normal? [23:32:10] <exobuzz> (not the api.php but the main code) [23:32:23] <wikibugs> 03(mod) Global deleted image review for Commons admins - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14801 (10brion) [23:32:34] <^demon> brion: Fair enough on {{apiurl}}. Question though: Mind me putting $wgApiScript back in? As it stands, there's no definite way to construct the URL to the API except by using $wgScriptPath and assuming it's called api.php [23:33:05] <brion> ^demon: well that's what the entire system thinks, so it better be right :) [23:33:19] <^demon> True ;-) [23:33:21] <brion> otherwise we're making wrong links to our own api when attempting to use it (eg for opensearch) [23:33:27] <CIA-55> 03simetrical * r38264 10/trunk/phase3/maintenance/parserTests.txt: Add twelve new parser tests, of which four fail, to cover bug 6200. Brings failing tests back up to 15 for me. [23:34:40] <wikibugs> 03(mod) Linebreaks are mishandled in <blockquote> and <li> - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=6200 +comment (10Simetrical+wikibugs) [23:35:18] <ryno1> brion: Any advice on how to find out what the problem is? I am relatively new to MediaWiki... [23:35:51] <^demon> brion: Oh btw, I commented on the IPv6 bug. I was playing with the installer earlier and the IPv6 check in defaultsettings caused a regression there, as the IP class obviously isn't loaded during install :-P [23:36:00] <ryno1> maybe a way to bypass the loadbalancer all together [23:36:02] <wikibugs> 03(mod) Page URLs are messed up when using accented letters (non-ascii) mixed with escaped characters. - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=11097 (10brion) [23:36:17] <brion> ryno1: check your configuration, check the db, check the manual changes you made to the system [23:36:34] <ryno1> Thx brion [23:36:49] <brion> ^demon: well, the installer shouldn't be running on our site o_O :) so there must be something else somewhere..... [23:37:40] <^demon> Well true, but it made running the installer impossible, which affects pretty much every MediaWiki install except Wikimedia's :-P [23:37:57] <wikibugs> 14(INVALID) default uselang problems - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14802 +comment (10brion) [23:38:28] <brion> :D [23:39:10] <exobuzz> brion: Category:Wikipedia:Translation does work ok with that change it seems. [23:39:23] <brion> yay [23:39:26] <brion> hmm [23:39:36] <brion> well on your local that's probably not a namespace [23:39:47] <brion> try Category:Project:Translation [23:39:52] <brion> well [23:40:01] <brion> that'll work -- try putting it in as just "Project:Translation" [23:40:32] <exobuzz> aah ok [23:40:37] <exobuzz> will test [23:41:08] <^demon> I was playing with that earlier. Nice little fun bug. I eventually figured it would probably be easier to just str_replace() the namespace off the front of it if it exists ;-) [23:41:15] <CIA-55> 03jhsoby * r38265 10/trunk/ (7 files in 3 dirs): (log message trimmed) [23:41:15] <CIA-55> Adding accesskey="s" to forms in the following special pages: [23:41:15] <CIA-55> * Special:BlockIP [23:41:15] <CIA-55> * Special:EmailUser [23:41:15] <CIA-55> * Special:ExpandTemplates [23:41:16] <CIA-55> * Special:Export [23:41:18] <CIA-55> * Special:Import [23:41:26] <brion> ^demon: but there might be 50 aliases :) [23:41:31] <brion> and different case variants [23:44:56] <wikibugs> 03(FIXED) Logo for ext.wikipedia - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14804 +comment (10brion) [23:45:47] <wikibugs> 03(mod) Install News Channel extension on Wikinews - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14805 (10brion) [23:45:51] <^demon> Bah, I say. [23:46:26] <exobuzz> brion: i see - you are right. so maybe it is just easiest to drop Category from the beginning if its there and use the old line ? [23:48:01] <brion> dunno, dunno :( [23:48:44] <exobuzz> unless there is some other api call ? [23:48:45] <brion> perhaps run through newFromTitle *without* the default NS, then check the result NS [23:48:55] <brion> if it's not NS_CATEGORY, then run it through the makeTitleSafe [23:49:16] <^demon> brion: ForeignApiRepo needs a new name. That's 2 bugs in like 3 days that have assigned it to the API component of MW :-p [23:49:21] <wikibugs> 03(mod) Enable GlobalBlocking extension - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=8707 (10wiki) [23:49:31] <OverlordQ> Awww, parserTests only works for MySQL. Oh well, dont really need to run them anyways :D [23:49:39] <brion> alternatively we could just do the newFromTitle w/ Category default and say "use the full name if you rname is ambiguous, like with templates" [23:49:59] <brion> OverlordQ: sounds like an opportunity to make it portable! ;) [23:50:13] <brion> ^demon: >:D [23:50:16] <brion> maybe RemoteRepo [23:50:39] <^demon> Bah, then someone will wonder why it doesn't turn on their TV.... [23:50:53] <brion> if they had the right api it could! [23:50:58] <^demon> hahaha [23:53:12] <OverlordQ> >_< [23:53:16] *OverlordQ goes function grepping [23:55:04] <^demon> grepping is always fun. especially when it's something incredibly vague. [23:55:42] <brion> grep -ri "why doesn't my code work" phase3/includes [23:56:07] <wikibugs> 03(mod) Microsoft SQL Server/MSSQL support (tracking) - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=9767 (10dj.bauch) [23:57:07] <^demon> brion: grepping for fixme/todo is always fun too :-) [23:57:29] <OverlordQ> is there a different way to find out what a db handle backend is besides looking at $db->getSoftwareLink? [23:59:02] <brion> OverlordQ: well, the class name of the db object should give you a clue if you're just looking for which db type :) [23:59:13] <Ryanposs> Could anyone look at my en.wiki main page desgin? I've got a few coding problems with it that I need fixing [23:59:39] <OverlordQ> oh wait, Im an idiot [23:59:40] <OverlordQ> $wgDBtype [23:59:42] <wikibugs> 03(mod) AntiSpoof should return more than one result - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=12232 (10brion)