[00:01:03] TheLetterX: Also, next time, include RELEASE-NOTES in your patch [00:01:10] Ok [00:02:05] TheLetterX: Are you already in the CREDITS file? If not, how would you like to be credited? [00:02:16] Hang on [00:02:33] is there a hack to import utf-8 images with the maintinence/importImages.php script? [00:02:41] "Soxred93" [00:02:44] because right now it chokes on images with a utf-8 title [00:02:57] OK [00:02:58] TheLetterX is used only on enwiki [00:04:01] Yay, RELEASE-NOTES conflict :@ [00:04:17] Hah [00:04:58] RoanKattouw, we need to have a special merge script that will correctly merge release note conflicts on svn up. [00:05:09] I know git and hg and so on support custom merge scripts, I wonder if svn does too? [00:05:09] 03catrope * r40116 10/trunk/phase3/ (5 files in 4 dirs): (bug 13471) Added NUMBERINGROUP magic word. Patch by Soxred93 [00:05:17] Simetrical: Well I've never seen an incorrectly merged RELEASE-NOTES [00:05:34] You mean auto-resolve RELEASE-NOTES conflict when SVN fails to? [00:05:50] RoanKattouw, yes. [00:05:51] Like when it says Conflicted instead of Merged [00:05:58] You should use the merge algorithm "theirs, then mine". [00:06:09] where'd he go? [00:06:11] Which is a really bad idea for code, but exactly what we want for release notes. [00:06:20] 03(FIXED) Add NUMBEROFROLLBACKERS magic word - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13471 +comment (10roan.kattouw) [00:06:34] Simetrical: Maybe apply that to RELEASE-NOTES only? [00:06:44] Code conflicts should *never* be auto-resolved [00:06:59] RoanKattouw, code conflicts are always auto-resolved, if they're on different lines. :) [00:07:11] But yes, of course this would be RELEASE-NOTES only, as I said. [00:07:12] I mean real conflicts [00:07:34] Well, then by "real" you mean "conflicts that haven't been automatically merged". [00:07:43] Anyway, a custom merge script could discriminate between files and so on. [00:07:46] As in conflicts where two people try to change the same line [00:07:56] rather than changing different functions that just happen to be in the same file [00:09:25] am i the only one who sees issues with that? [00:09:36] Xiong: Issues with what? [00:10:17] if two devs change two different lines in the same function, there is no guarantee they won't screw each other up [00:10:26] correct [00:10:28] That's true [00:10:43] They could screw each other up if they change two different lines in entirely different files. [00:10:49] i don't see how any sort of auto-merge could not blow everything up [00:10:51] In practice, it's rare enough that auto-merging is safe. [00:11:03] But in the svn up output (or in the TortoiseSVN dialog) you'd see that your file was merged [00:11:03] Xiong, because usually people aren't changing around things in a way that conflict. [00:11:14] That's also true [00:11:22] well at least in theory, the interface to a function should be well-defined, so the internals can change at will, independent of other work [00:11:25] People aren't gonna change stuff that breaks something somewhere else [00:11:30] It does very rarely blow up when changes are incorrectly merged, but usually in that case you have a conflict on at least one line anyway. And if not, you can fix it after the fact. [00:11:44] Xiong, but people can change interfaces to functions. [00:11:48] Code refactoring and all. [00:12:00] well that's risky business [00:12:06] In which case they'd change all function calls too and probably conflict with you [00:12:21] the "further away" the function call is from the function definition, the riskier [00:12:22] Xiong: Not necessarily. It doesn't happen often, but it happens from time to time in MW [00:13:25] anyway, if a fellow changes the interface to a function, it's his responsibility to change all calls to it before commit [00:13:34] RoanKattouw, you could be adding a new function call and miss their changes to the old ones. [00:13:36] It can happen. [00:13:46] You would never change an interface to a function. [00:13:50] Yeah, that's possible. But it doesn't happen very often [00:13:57] Xiong, yes, but that's the point, we're talking about someone changing a function at the same time someone else adds a new call to the function. [00:14:01] ME, i'm kind of in your camp [00:14:05] RoanKattouw, right, that's my point. [00:14:16] Auto-merging is fine as long as it works in almost all cases. [00:14:21] Doesn't have to be perfect. [00:14:25] i thought the thing on the table was merging changes made to the same function [00:14:27] But refactorings are rare, adding function calls is not common, so this weird kind of conflict is very rare [00:14:59] Xiong: At first, yes. But then we decided that since you're not gonna change something that breaks stuff elsewhere in the function, you can't really break newly added code either [00:15:06] one function, i change line 5, you change line 8, the auto-merger stitches together both changes [00:15:54] Yeah, but line 5 breaking line 8 is very rare [00:15:59] you might be fixing something here while some other guy is fixing the same thing there [00:16:04] say, an array index issue [00:16:17] In which case it probably does no harm [00:16:24] i fix it here by writing $i+1, you fix it there by writing $i-1 [00:16:35] which means that the issue is still broken [00:16:43] That's true [00:16:56] But you'd notice that because of the RELEASE-NOTES conflict [00:17:11] i run into a lot of that stuff in my own code, although i try hard to avoid it [00:17:18] Because you probably didn't word it exactly the same [00:17:27] i wind up in a corner where i'm using two yardsticks to measure the same thing [00:17:28] And if you don't notice it there, you'll notice it when marking the bug as fixed [00:17:36] Which it already has been by the other guy [00:17:40] but one has a zero at 0 and the other starts at 1 [00:18:05] well, i'm foolish to argue the point as if i had any secret knowledge [00:18:19] i've voiced my opinion, it gives me the willies, that's all [00:18:40] Yeah sometimes it is scary [00:18:53] But like I said, you'll notice when svn up says something's been merged [00:19:18] well does that include a rubber slapper so that both devs go and check the merged version? [00:19:36] Okay, look. [00:19:45] sorry, this really is a style issue, nothing else [00:19:55] Fact of the matter is, every RCS automatically merges changes not to the same line. [00:19:56] i like workers to check out one file, fool with it, and put it back [00:19:59] It works fine. [00:19:59] End of story. [00:20:21] Xiong: Dunno if you know, but the merge happens *before* you commit, you the second guy double checks that the merge didn't break anything *before* he commits his stuff [00:20:32] ah [00:20:36] well now that is fine [00:20:36] And what Simetrical said. Theory != practice [00:20:45] i completely remove all objections [00:21:00] so long as the onus is on the second guy before he hits the button, no problem [00:21:02] sorry [00:21:22] may even save time [00:23:17] OK guys it's 2:20 AM over here, so I'm gonna get some sleep now [00:24:20] it's 5:30pm here and brutally hot, too hot to type or read [00:24:27] so i'm gonna zone out now [00:45:04] Morning all! I wonder if I could get some help. I entered a feature enhancement request on the 'zilla, and its gone. Completely. I think it was 15339 or something like that. Can anyhone help? I have had no email for it... [00:45:53] found it. its in my log. [00:45:56] (NEW) Reference creation button produces incomplete entries in extended toolbar. - https://bugzilla.wikimedia.org/show_bug.cgi?id=15339 enhancement; normal; MediaWiki: General/Unknown; (hammer.of.thor) [00:46:16] sorry for the paste, just referring to the bug that went walkies [00:55:10] 03nad * r40117 10/trunk/extensions/SimpleSecurity/SimpleSecurity.php: warnings were being generated when php using E_ALL [01:09:55] 03demon * r40118 10/trunk/phase3/includes/SiteStats.php: Add some caching to the group counts. [01:10:37] <^demon> Simetrical: There's the caching you mentioned on wikitech-l. Group stat checks are now cached for an hour. [01:10:56] ^demon, ah, of course. [01:11:02] Oh, not of course. [01:11:06] Anyway it works. [01:11:19] You might want to increment it/decrement it when people are added to/removed from the group. [01:11:31] Or just delete it from the cache. [01:11:50] <^demon> Let me look at that. [01:20:16] <^demon> Simetrical: Can you ping me tomorrow about it? [01:20:42] 14(WFM) DISPLAYTITLE doesn' t always work when current revision is sighted - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=15330 (10JSchulz_4587) [01:20:55] ^demon, it doesn't really matter. [01:21:17] <^demon> Alright then. Have a good one. [03:09:53] domas: why is $wgQueryCacheLimit set so low? [03:36:29] 03nad * r40119 10/trunk/extensions/SimpleSecurity/SimpleSecurity.php: Die if MediaWiki version < 1.11 [03:52:49] 03(ASSIGNED) Go button on Special:RecentChanges and Special:Watchlist - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=15172 +comment (10soxred93) [03:53:38] I'm being productive :) Could someone please review it? [03:56:23] Actually, would it be allright if I were to get SVN accesS? [03:56:53] TheLetterX: you'll need to ask brion when he comes on [03:57:00] Urgh [03:57:43] brion won't be on much at all this week [03:58:38] Tim Starling occasionally grants access as well [03:58:59] Neither are on *facepalm* [03:59:20] Tim is, but he's away [04:02:07] 03(mod) Go button on Special:RecentChanges and Special:Watchlist - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=15172 +comment (10soxred93) [04:02:14] Ah, I see [04:02:20] yeah and anyway you'll have to wait a bit [04:03:01] usually svn accounts are created by batch, every month or so [04:19:37] <[Hyarion]> hmm, I don't suppose there's any extension which adds something like "UserXYZ is currently editing this article" when someone edits an article currently being edited? [04:21:43] I think this page is done with a custom parser function: http://www.archiplanet.org/wiki/Add_A_Page [04:21:51] it's VERY cool [04:22:08] would anyone know how I might go about emulating it [04:22:10] ? [04:24:08] does anyone know why user names are capitalized on both first name and second name on Wikipedia? [04:24:55] or not why so much as how it can be done [04:26:05] jetole: i think simetrical mentioned to talk to me the other day? [04:26:15] um [04:26:16] yes [04:26:18] about ldap [04:26:21] jetole: ah [04:26:23] wtf was my question [04:26:26] jetole: heh [04:26:30] I have ldap working [04:26:35] so perhaps it has passed [04:26:42] however I have another question [04:26:42] ah. ok. thats cool [04:26:45] ? [04:27:14] usersnames start with a capital because they have to be valid page names, to answer your last question [04:27:21] can I make it so someone logs in with their AD domain login name, in my case jetole, and it displays something else for my user name such as the part of AD that lists me as Joseph Tole [04:27:25] ? [04:27:39] depends on what you mean [04:27:57] you can pull preferences, and have the real name populated from a field [04:28:15] the newest version of the plugin even lets you chose the attribute [04:28:26] Ryan_Lane: well, right now, I am choosing to either login as jetole or Joseph Tole but would like to login as jetole and have it display "Joseph Tole" as the User: [04:28:35] ah ok [04:28:52] it *is* possible, but it may be a bad idea [04:28:59] Joseph Tole seems like an odd way to login but I don't want our comapny wiki... [04:29:02] fair enough [04:29:10] we will stick with it this way then [04:29:14] unless you know for sure the attribute you want to use is unique [04:29:15] now the other thing I just asked you [04:29:25] ok, I will wait on that [04:29:47] yes it is, for example I want to login with jetole and have Joseph Tole displayed, it is our internal company AD [04:29:52] so they are unique names [04:29:55] but, if you want user's signatures to show their name instead of the username, that is totally doable [04:30:05] No [04:30:13] not sure... [04:30:23] well, what if you get two employees that have the same first and last name [04:30:36] ? [04:30:40] I would like to have people login with their login name, i.e. jetole, but be able to pull User:Joseph Tole [04:30:55] right, but that page is the username [04:30:56] I will address that when it happens? [04:31:05] ok, fair enough [04:31:07] ok forget about it then [04:31:08] lemme show you how to do it [04:31:30] i just wanted to give you fair warning first ;) [04:31:54] Well I am senior net admin and our firm is 30+ people so I don't think too much bad can be expected [04:32:01] I manage AD and wiki [04:32:04] ah ok [04:32:13] sounds like uniqueness is doable [04:32:13] you think it sounds fair or should I pass? [04:32:23] ok [04:32:28] http://www.mediawiki.org/wiki/Extension:LDAP_Authentication#Smartcard_authentication_options [04:32:39] ignore the fact that it says smartcard [04:32:46] what you want is the function and the hook [04:32:47] already done, lol [04:33:14] this line inparticular: [04:33:16] $LDAPUsername = $info[0]['samaccountname'][0]; [04:33:27] 03(NEW) Links error in Chinese Wikipedia - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=15344 04BLOCKER; normal; MediaWiki: Page rendering; (PhiLiP.NPC) [04:33:36] I'm not sure what attribute you want to pull, but it isn't likely to be samaccountname [04:33:39] yeah [04:33:41] I see that [04:33:51] it'll probably be cn or displayname [04:34:01] no but I will look it up tomorrow, I am saving this to google bookmarks atm [04:34:04] *jetole is at home [04:34:08] ah [04:34:10] yes [04:34:11] Can anyone tell me anything about http://www.archiplanet.org/wiki/Add_A_Page ? [04:34:21] Ryan [04:34:23] thank you [04:34:33] let me make sure that function is called normally [04:35:01] I want to truncate everything in the article and just use the hook and function? [04:35:10] everything above that I ignore, right? [04:36:08] yeah [04:36:18] well, you want the rest of your normal configuration [04:36:29] but ignore the other smartcard stuff [04:36:34] right, what I have now [04:36:44] and add the hook and func [04:37:01] yeah [04:37:06] awesom [04:37:10] awesome too [04:37:14] :D [04:37:16] *MZMcBride points Tim-away to bug 15344. [04:37:20] now for capitalization... [04:37:56] on Wikipedia I can have an articale written by Joseph Tole but in mediawiki, right now we have articals that ARE written by Joseph tole [04:38:11] I saw [04:38:27] mediawiki is case sensitive for articles [04:39:17] jetole: you can rename the user and it'll make the username show properly in recent changes, history and such [04:39:21] Ryan, but our user names are defaulting the last name to lower case [04:39:48] hmm. I bet the ldap plugin is doing that [04:39:52] even though the name isn't [04:40:02] really? I didn't think so [04:40:07] I just thought I would ask [04:40:08] although, it shouldn't unless you tell it to [04:40:18] is that how the user is logging in? [04:40:22] I thought it was a mediawiki thing by default [04:40:26] no [04:40:36] it is doing this to all users and articles [04:40:59] everything except the first letter is lowercase? [04:40:59] it is doing this to all users _and_articles_ [04:41:11] which is why I didn't think ldap [04:41:13] yeah [04:41:14] that is kind of weird... [04:41:20] I am only asking you because you are here [04:41:28] I don't think it's your "department" [04:42:38] jetole: ah, btw, what type of ldap config are you doing? straight binds using USER-NAME@DOMAIN? [04:43:11] jetole: for the hook/function thing to work, you'll need to do search based binds [04:43:16] straight binds but not sure about the syntax [04:43:21] should I ssh in and check [04:43:35] you mean with a default user name that logs in? [04:43:48] *jetole ssh's [04:44:23] ack. let me retract that [04:45:05] 'XXXDomain' => 'CN=USER-NAME,CN=Users,DC=XXX,DC=officedomain' [04:45:15] i forgot that I put something special in for AD style straight binds. you'll still need to add a basedn and possibly a search string [04:45:16] oh [04:45:29] that is for ... [04:45:33] $wgLDAPSearchStrings = array( [04:45:49] use 'XXXDomain' => 'USER-NAME@DOMAIN' it works and is simpler [04:45:57] with AD? [04:46:02] oh [04:46:03] right [04:46:09] I misread that at first [04:46:27] Dantman: Ping. [04:46:41] Hmm? [04:46:46] hold a moment and I will sed out the sensitive (company name which I can't share) and paste the whole ldap config [04:46:57] The new
above the edit form looks absurd. Please, please revert (or remove). [04:47:00] use pastebin :) [04:47:56] Huh? [04:47:57] was about to [04:48:15] http://pastebin.com/m7e032f00 [04:48:21] Dantman: It was your change, no? The

is now very visible. [04:48:28] MZMcBride: That has been there forever... I just moved it from inside previout to outside... [04:48:36] Ryan_Lane: CompanyName/companyname was sed out and that is all [04:48:57] Dantman: Well, it added about a centimeter of whitespace, which looks really goofy. [04:48:57] hmmm [04:49:14] Ryan_Lane: AdminDomain was kept because jetole was set as the sysop before LDAP [04:49:45] jetole: when using @, you probably want to use the short uppercase name, so USER-NAME@COMPANYNAME [04:49:54] How about I put that note on a todo list, and get on with migrating to a margin tomorrow? [04:49:55] also, for capitalization, O think this will work => http://www.mediawiki.org/wiki/Manual:$wgCapitalLinks [04:49:55] unless what you have is working [04:50:15] what I have works [04:50:20] ok. thats cool [04:50:24] I've been thinking of elimitating that html cruft in favor of CSS [04:50:55] like I said, just having to login with "Joseph Tole" instead of "jetole" just to have our proper names appear is a little, well, just doesn't seem right [04:51:04] jetole: you probably also want to use the same syntax for the companydomain [04:51:13] ok [04:51:16] it is more flexible [04:51:32] well if I am changing back to that format then I can lose admindomain entirely [04:51:37] that is one thing i think microsoft actually got right :) [04:51:39] it was kept so jetole could login [04:51:51] MS got something right? [04:51:56] ;) [04:52:10] Dantman: That would be much better. A simple padding on one of the tags would more than suffice. [04:52:12] *jetole <= heterogenous admin [04:52:30] we run mediawiki on ubuntu and auth against AD [04:52:34] jetole: you'll probably need to add the proxyagent back in [04:52:49] ok [04:52:50] jetole: and the password, and the basedn [04:52:58] oh [04:53:06] so straight bind will not work [04:53:08] jetole: because it'll need that to pull the user's info [04:53:16] jetole: no, it'll work [04:53:18] I see [04:53:22] jetole: this is like a mix of the two [04:53:28] ok [04:53:32] 03(mod) Links error in Chinese Wikipedia - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=15344 normal->high (10klneast) [04:53:36] jetole: if you really want to, you can switch over to search based completely [04:54:02] jetole: make a proxy user and strip all of its rights; only allow it to bind and search [04:54:25] ok [04:54:46] and yes, microsoft did fairly well with AD :) eventhough i curse its name at times [04:54:55] well it seems I have enough to play with tomorrow [04:55:01] hehe [04:55:04] I do at SSO [04:55:16] it's a pain in the ass to have it replicate with non AD [04:55:27] multi master I mean [04:55:28] try adding smartcards into the loop, and then add another directory server in [04:55:52] well, we have two [04:55:56] two AD I mean [04:56:02] servers [04:56:03] then consider kerberos, and realize you have to switch everything over to AD :( [04:56:15] hehe [04:56:38] do you have multiple domains, or just one? [04:56:43] I would like to be able to have a linux server where I can add a user to the ldap and have a windows pro station which auths against the AD be able to use that user [04:56:53] or use a linux server as the AD server [04:56:58] two [04:57:03] jetole: samba [04:57:14] jetole: and samba v4 will be able to do win2003 domains [04:57:24] from what I understand samba can only do this via w2k PDC [04:57:29] wha [04:57:31] really? [04:57:34] yep [04:57:56] and you can add kerberos and an ldap server as a backend [04:58:00] I read in a couple places samba only does w2k PDC but it may be deprecated [04:58:09] samba v4 isn't out yet [04:58:12] I mean what I read may be deprecated [04:58:15] oh [04:58:18] svn? [04:58:32] yeah, it may be beta now? i think? i'm probably wrong :) [04:58:41] *jetole looks [04:59:10] still alpha as of june [04:59:41] hmmm [04:59:43] oh well [04:59:50] the fact that it will do w2k3 is great [05:00:12] heh, wonder what that means about w2k8 but we don't use it _YET_ [05:00:22] yep, until microsoft releases AD2008 [05:00:30] and makes all of its clients require it [05:00:46] so there is no AD2008? [05:00:53] *jetole sighs in relief [05:01:08] anyway, i'm off to bed. good luck with the plugin, and if you need any help, let me know via email, wiki, or here (although i'm usually only here after hours) [05:01:32] I appreciate all the help you have given [05:01:38] n/p [05:01:40] see ya [05:01:42] and I am off to bed too [05:01:46] cheers [05:08:45] 03(mod) Links error in Chinese Wikipedia - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=15344 +comment (10PhiLiP.NPC) [05:18:31] AaronSchulz: http://de.wikipedia.org/w/index.php?title=Tel_Aviv-Jaffa&oldid=49632438&diff=cur&diffonly=0 shows a few unsighted images but these images are all on Commons... how to sighten this? [05:22:30] 03tstarling * r40120 10/trunk/phase3/includes/parser/LinkHolderArray.php: [05:22:30] * Fix bug 15344. Was caused by the overwrite of $entry, assigned by reference to the last link and then overwritten in later code. [05:22:30] * Split doVariants() out from replaceInternal(), big functions are prone to variable reuse bugs [05:23:32] is there a spec on mw's anchor encoding? [05:24:30] 03(FIXED) Links error in Chinese Wikipedia - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=15344 +comment (10tstarling) [05:26:49] jeremyb: yes [05:26:50] ahah jeremyb :) [05:27:02] the spec is written in the form of PHP code [05:27:08] NicDumZ: sorry about the double spam :) [05:27:20] at line 788 of includes/Sanitizer.php [05:27:51] well, I think your patch adresses the issue anyway [05:28:50] 03(mod) Transfer some Features from DPL to - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13151 +comment (10naught101) [05:29:12] Tim-away: i'm looking at http://en.wikipedia.org/w/index.php?title=\langle&action=history and am puzzled [05:29:17] NicDumZ: different bug [05:29:41] NicDumZ: https://sourceforge.net/tracker/index.php?func=detail&aid=2058951&group_id=93107&atid=603138 [05:32:42] hrmm [05:33:15] NicDumZ: looking at http://svn.wikimedia.org/viewvc/mediawiki/trunk/phase3/includes/Sanitizer.php?view=annotate#l788 i'm wondering how we should detect what $flags is [05:35:09] hi, all. question: in my local wiki (en.wiktionary) we have patrolling for all diffs, not just newpages. if someone were to have the ability to patrol edits, then would all his own edits be marked patrolled automatically? [05:35:30] jeremyb: INITIAL_NONLETTER [05:35:48] 03(mod) {{PAGELENGTH}} magic word - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13483 +comment (10naught101) [05:36:00] TimStarling: callers can pass it explicitly right? [05:36:15] you asked about anchor encoding, right? [05:36:17] (otherwise why test for it?) [05:36:24] hello all! I need help with something [05:36:57] TimStarling: yes, i was looking at the history i linked (via the pywikipedia bug i linked) [05:36:57] im using php5 and my wiki is at wiki.kiaotiongster.org [05:37:22] jeremyb, I grepped for the calls, as of now _all_ calls do not define $flags, just using the default value. I think we can assume it's always INITAL_NONLETTER [05:37:39] ho can i access the page without using index.php5 [05:37:41] NicDumZ: okey [05:38:32] NicDumZ: shall i add that? [05:38:40] how can i access the page without using index.php5?title=blah_blah to wiki.kiaotiongster.org/title [05:39:11] !shorturl | artagnan5 [05:39:11] --mwbot-- artagnan5: To create simple URLs (such as the /wiki/PAGENAME style URLs on Wikimedia sites), follow the instructions at . There are instructions for most different webserver setups. If you have problems getting the rewrite rules to work, see !rewriteproblem [05:39:24] yay, wild guess worked :) [05:40:05] yep jeremyb add it. btw #pywikipediabot is usually less crowded :p [05:40:31] NicDumZ: oh, i thought #wikipedia-bag was #pywikipediabot [05:40:58] hehe well no. hopefully all bots do not run pywp [05:41:01] :) [05:41:26] and bag is only for en.wikipedia ;) [05:42:02] question: in my local wiki (en.wiktionary) we have patrolling for all diffs, not just newpages. if someone were to have the ability to patrol edits, then would all his own edits be marked patrolled automatically? [05:43:30] msh210: fwiw, that's borderline #wikimedia-tech territory. but here *should* work too [05:44:38] thx, maybe i'll try there... [05:47:23] 03dale * r40121 10/trunk/extensions/MetavidWiki/maintenance/ (3 files): remove extra Speech By text in annotative layers [05:55:39] !rights [05:55:39] --mwbot-- For information on customizing user access, see < http://www.mediawiki.org/wiki/Help:User_rights >. For common examples of restricting access using both rights and extensions, see < http://www.mediawiki.org/wiki/Manual:Preventing_access >. [05:58:35] 03(mod) Links error in Chinese Wikipedia - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=15344 (10PhiLiP.NPC) [05:59:06] *MZMcBride finds this a bit worrying: http://www.mediawiki.org/wiki/Extension_namespace_registration [06:09:57] 03(FIXED) Lots of images have disappeared from ku.wikipedia.org - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=15111 -shell ; +comment (10raimond.spekking) [06:12:41] 03(mod) Lots of images have disappeared from ku.wikipedia.org - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=15111 (10raimond.spekking) [06:39:04] <_wooz> lo [07:14:46] 03raymond * r40122 10/trunk/phase3/ (RELEASE-NOTES includes/specials/SpecialRecentchanges.php): * (bug 15172) 'Go' button of Special:Recentchanges now on the same line as the last input element (like Special:Watchlist too) [07:17:15] 03(FIXED) Go button on Special:RecentChanges and Special:Watchlist - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=15172 +comment (10raimond.spekking) [07:19:56] !rar | Splarka [07:19:56] --mwbot-- Splarka: rar [07:54:41] Can somebody check their MediaWIki "LocalSettings.php" file! I see the file start with """ tags of some sort. That can't be good??? [07:56:56] apprentice: it's not required [07:57:03] ic [07:59:42] we used to use ?> [07:59:58] but people would accidentally put line breaks after them [08:00:05] which messed up the output [08:00:22] so a couple of people went through the whole repository and removed them all [08:05:14] so you've made the "?>" tag invisible. Theoretically it should be there? [08:06:35] apprentice, ?> does not have to be there [08:06:41]