is what made me think it was doing them inline [01:04:19] nothing like this on de.wikipedia. [01:05:14] eh, sorry. [01:05:54] you're right, there is when i'm not logged in. maybe because i disable all the onload-hooks in my monobook.js [01:07:23] hmm. just try it. i need sleep. good night! [01:13:03] 03(mod) On-page "What Redirects Here" for admins - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=11083 +comment (10kovo1) [01:17:41] 03(mod) don't advertise "create account" if not allowed - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=11177 (10jidanni) [01:33:14] Simetrical: http://svn.wikimedia.org/viewvc/mediawiki/trunk/phase3/RELEASE-NOTES?r1=25501&r2=25502&sortby=date [01:33:16] heh [01:34:35] that was a bit odd [01:48:02] how do you get fixed-width text boxes, for example if I want to have programming code exerpts on a page [01:55:17] like on http://en.wikipedia.org/wiki/Pseudo-code where below it says pseudocode: there is a text-box with fixed-width font inside [02:08:39] I just went through an extremely painful experience upgrading from PHP 5.2.1 to PHP 5.2.4 on the WIMP platform, including issues with php-cgi (the recommended approach from Windows Installation guides) and other issues. 2 questions, 1) Should I document this in MediaWiki? 2)If so, where? [02:42:28] Hi there. I'm having some problems transferring my local installation to a hosted server. For some reason my mainpage links are coming up as domain.com/index.php/Article. Once I get away from the main page its ok and continues with the usual ugly urls that I have set up. Any ideas where to start? I'm happy with ugly urls for the moment. [03:00:40] 03tstarling * r25516 10/trunk/extensions/OggHandler/OggHandler_body.php: -ss must be specified before -i, to enable fast seek. But see http://bugzilla.mplayerhq.hu/show_bug.cgi?id=892 . [03:04:03] 03(NEW) Edit to article attributed to wrong author - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=11188 major; high; MediaWiki: History/Diffs; (ldsw) [03:04:21] 03(mod) Edit to article attributed to wrong author - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=11188 (10N/A) [03:14:13] 03(mod) Edit to article attributed to wrong author - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=11188 (10N/A) [03:14:48] 03(mod) Edit to article attributed to wrong author - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=11188 (10N/A) [03:19:33] 03(mod) uploads on is:wikt only for sysops - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=11187 +shell (10shinjiman) [03:34:02] 03tstarling * r25517 10/trunk/extensions/OggHandler/ (README ffmpeg-bugfix.diff): About the FFmpeg bug. [03:39:29] 03kim * r25518 10/trunk/extensions/Wikidata/util/copy.php: More minor refactors (tidied) [04:10:46] 03(NEW) Search on http://wiktionary.org/ searches wikiquote rather than wiktionary. - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=11189 major; normal; Wikimedia: Site requests; (gmaxwell) [04:16:25] 03(mod) Search on http://wiktionary.org/ searches wikiquote rather than wiktionary. - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=11189 normal->high; +shell (10cannon.danielc) [04:24:43] 03(FIXED) Search on http://wiktionary.org/ searches wikiquote rather than wiktionary. - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=11189 +comment (10cannon.danielc) [04:26:41] 03(mod) Add section edit link for 0th section - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=156 (10circee) [05:31:31] Heya... [05:31:50] All of a sudden my MW died today and gives me this error: Fatal error: Allowed memory size of 8388608 bytes exhausted (tried to allocate 1245184 bytes) in /home/path/html/languages/messages/MessagesEn.php on line 2040 [05:31:56] o_O [05:33:18] I have a string of the form xx.yyy.z, and I'm trying to match yyy (the next-to-last element separated by dots). This is not working: [05:33:21] /.*([^.]+)\.[^.]+$/ [05:35:03] why not using greedy modifier [05:35:21] /.*?\.(.*?)\.(.*?)/ [05:36:58] Jasio: 8 MB is too little [05:37:03] Nope, that doesn't work either, puzzlet. [05:37:20] What my original regexp matches is just one character before the last dot, instead of the entire non-dot span before it. [05:38:58] why do you have .* in beginning? [05:39:04] hey all [05:39:20] >>> re.match(r".*?\.(.*?)\..*?", "xx.yyy.z").group(1) [05:39:20] 'yyy' [05:39:51] quick question, does anyone know of mediawiki markup to HTML converters that actually work? php/python/ruby prefered [05:40:06] sour1: hehe. [05:40:09] parser.php :D [05:40:10] placing ? after * makes it greedy match [05:40:21] I thought that made it a lazy match. [05:40:38] Nikerabbit, if it's that easy, pass me a rifle - i've spent almost an entire day of works time trying to figure where mediawiki keeps it :P [05:40:52] sour1: Best of luck! [05:41:35] sour1: it's the only that fullfills the requirement "actually works" :D [05:42:09] Nikerabbit, funny thing is, I've never had it pop up before. [05:42:15] It came up today. [05:42:15] .*? matches as least characters as possible to confort rest of the regexp [05:42:31] haha, notepad++ crashes when i try open parser.php [05:42:34] well it's beyond my English skill [05:43:27] Jasio: then the limit is not working anymore [05:43:56] puzzlet: your patterns are matching nothing here. [05:44:02] sour1: use of notepad is discouraged for other reasons [05:44:49] it's notepad++, "enhanced notepad", i like it cause it's something lightweight when working under windows [05:44:59] Use SciTE. [05:45:05] Notepad doesn't handle UTF-8 well. [05:45:09] davidmccabe: what programming language are you using [05:45:18] davidmccabe: does that require to escape \'s? [05:45:26] php / pcre. [05:45:30] And doesn't Notepad add BOM's which PHP chokes on? [05:45:31] oh, the answer is: [05:45:36] /.*?([^.]+)\.[^.]+$/ [05:45:43] which makes just that initial match lazy. [05:46:16] jasio, notepad++, not windows notepad [05:46:22] SciTE is as "light weight" as you can get. [05:46:32] No installer, entire application is wrapped in a 400kB exe. [05:46:41] nice, i'll look into it now [05:46:47] thanks for the info [05:47:15] Features language highlighting, UTF-8 support, and all those other magical features people crave for but never really use :p [05:47:18] can it handle large files well? 100mb+? [05:47:22] Sure. [05:50:00] you're right, very cut down :P, i'll test it against large files next time [05:50:44] I love it since I can pop it on a USB stick and take it anywhere. [06:08:45] 03siebrand * r25519 10/trunk/phase3/languages/messages/MessagesFrc.php: Fallback language was supposed to be English (removed French) [06:10:30] my head is begining to hurt [06:11:04] we're sorry. [06:11:08] <_wooz> lo [06:17:08] ok thanks for your help Nikerabbit, i think i can work it all out from here [06:26:23] uga [06:27:47] no robchurch here lately [06:27:47] mm [06:40:40] mm [06:47:45] Rob wants to fight me back! [06:48:00] haha! [06:48:02] committed deathmatch [06:48:06] s/ed/er/ [06:48:28] 03(NEW) Suggestion - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=11190 15enhancement; normal; Wikimedia: Bugzilla; (ekatz) [06:48:29] flyingparchment: I just don't get the idea of additional indexes on recentchanges just because of minor features [06:49:00] stop the bloat! [06:49:40] "Suggestion" [06:51:12] maybe I should just be bold and revert stuff :) [06:56:11] 03david * r25520 10/branches/liquidthreads/extensions/LqtModel.php: optimization: creating a thread tree should now be linear; profiling shows some benifit on the data I have here. [06:57:50] which stuff? [06:58:40] Nikerabbit: 24607 - [06:59:58] um, in svn, what's the easiest way to undo a commit you just made? [07:00:44] davidmccabe: svn diff -r1:2 | patch [07:00:46] =) [07:02:51] 03david * r25521 10/branches/liquidthreads/extensions/LqtModel.php: oops, that was broken. undid last commit. [07:02:57] thanks domas. [07:03:38] 03(NEW) An Edit link for the "in other languages" box in the left of wiki page. - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=11191 15enhancement; normal; MediaWiki: Internationalization; (lapsap7+mwiki) [07:07:29] domas: it checked if a diff was in RC, diffs that are non-ajecent or too old aren't, hmm [07:08:10] domas: maybe you could just check for adjacency and give a message when one 'patrols' a stale edit thats not in IC [07:08:11] *RC [07:08:25] if really wanted to be a miser on the indexes [07:12:56] Can sb tell me a good mysqldump command for taking some imported xml files from mediawiki.org into my wiki database, please? [07:15:08] 03david * r25522 10/branches/liquidthreads/extensions/LqtModel.php: all fixed :) [07:16:24] 03david * r25523 10/branches/liquidthreads/extensions/LqtModel.php: (m) cosmetic [07:17:09] domas: gawd, actually maybe the patrolling index should be rc_patrolled,this_oldid,last_oldid [07:17:23] then you could filter RC to edits that are not patrolled and do that [07:17:36] which would at least justify the index more [07:23:50] VoiceOfAll: it simply can use 'namespace,title' index [07:24:13] mmm [07:24:18] *domas looking at edge cases [07:25:39] heya, i am using Parser to try convert Wiki markup to HTML, I've encountered a problem, Fatal error: Call to a member function getMaxIncludeSize() on a non-object in C:\wamp\www\w\includes\Parser.php on line 2785 [07:25:42] what the heck: http://en.wikipedia.org/w/index.php?title=User:ST47/BAGLU&action=history [07:27:49] i presume all i need to do is call webstart.php and wiki.php, create a Parser() object, then $ParserObj->internalParse ($contents); [07:28:13] Can sb tell me a good mysqldump command for taking some imported xml files from mediawiki.org into my wiki database, please? [07:28:35] Soroush83 use xml2sql [07:29:04] then import it as usual, make sur eyou have mediawiki installed with all tables in place before you do it. [07:29:37] sour1, is it a free software? [07:29:44] mmm [07:29:48] Soroush83 yepp let me find it for you [07:30:14] now I wonder, didn't I clean the recentchanges once already [07:30:19] it is still at one month [07:30:33] thanks very much ,does it accept prefix for database [07:30:53] it just dumps out an import file in MySQL format [07:31:21] then you can import it using; mysql -u blah -p blah databasename < largedump.sql [07:32:26] sour1, It's excelent in that case. I've been using a pywikipediabot program to do that. It sometimes took several days for some imports [07:33:22] Soroush83 another tip, when importing it, change the database type to MyISAM from inddob or whatever it is [07:33:39] went from 15 hours to 1 hour on a p4 [07:35:37] Soroush83 the file is namex xml2sql, i found it on wikipedia website a while back [07:36:37] http://meta.wikimedia.org/wiki/Xml2sql [07:37:02] that's what i've been using [07:37:15] thanks very much, getting and then test it to see how it is [07:37:30] domas: I was thinking about ns,title [07:37:53] the upper end cases (like GWB) may have like 70-100 edits [07:37:55] My connection is usually poor and I need to have some wikis at home :) [07:38:05] not so bad if done on submit only [07:38:20] well 'submit', it's not a true submit [07:38:24] Soroush83 i know the feeling :) [07:38:44] domas: still, wouldn't it be helpful to filter by patrolled or not? [07:39:20] oh, there already is a filter [07:39:29] how is it working then? [07:42:22] http://en.wikipedia.org/w/index.php?title=Hurricane_Felix_%282007%29&limit=500&action=history [07:42:31] 3 approximate day(s) of edits [07:42:39] 121.206 edit(s) per day [07:42:54] domas: or higher ;) [07:42:55] VoiceOfAll: the query is done just on difference page view [07:43:15] pity we don't have namespace,title,timestamp index [07:43:18] Cool, i've commented out about 30 lines out of Parser.php and still getting errors woohoo! [07:43:19] probably that needs to be added [07:43:45] Is rel="copyright" a good idea if the Monobook theme and images are licenced under a different license? [07:44:49] domas: it used to just only show if rcid is given, what was bad about that? [07:45:42] namespace,title,timestamp is very nice for watchlists [07:46:46] http://bugzilla.wikimedia.org/show_bug.cgi?id=10793 [07:46:58] VoiceOfAll: it doesn't need separate index [07:47:02] for single link [07:47:07] yeah, probably [07:47:22] domas: high well does the 'hide patrolled' work? [07:47:31] *how [07:47:58] 'This would make [Mark as patrolled] links for diffs from the History or for combined diffs possible as well.' [07:48:05] heh, not sure that makes sense [07:49:29] VoiceOfAll: actually, even rc_timestamp index fully suits the task [07:50:02] I am going to cry. [07:50:43] sour1: as my Hhungarian ENG167 prof always says "Don't cry!" [07:50:48] *Hungarian [07:51:47] I wish, I have been trying to get mediawiki -> HTML converter working for an entire day now :/ [07:53:17] where do all the mediawiki developers hang out at, purhaps one of them is kind enough to help me out. [07:55:33] domas: maybe it should use cur_id (page_id). If a made a bunch of edits to a page and moved it [07:55:42] 03david * r25524 10/branches/liquidthreads/extensions/LqtModel.php: [07:55:42] Big savings by not querying for every Threads::withRoot invoked from the [07:55:42] protection hook. Possible regression: protecting a talkpage doesn't protect the [07:55:42] threads on it. But I reverted and still got this behavior, so maybe I remember [07:55:42] wrong that that worked. [07:55:50] the RC titles would still be the old ones [07:56:09] VoiceOfAll: title can be used from revision object [07:56:54] yes, but the revision's title may not match rc_ns,rc_title [07:57:18] why so? [07:57:34] I'm not looking at revision title! [07:57:42] changed for now conds to: [07:57:42] array( [07:57:42] 'rc_namespace' => $this->mNewRev->getTitle()->getNamespace(), [07:57:42] 'rc_title' => $this->mNewRev->getTitle()->getDBkey(), [07:57:42] 'rc_timestamp' => $this->mNewRev->getTimestamp(), [07:57:43] 'rc_this_oldid' => $this->mNewid, [07:57:47] 'rc_last_oldid' => $this->mOldid, [07:57:49] 'rc_patrolled' => 0, [07:57:51] ), [07:58:05] tell me how it can be wrong and I'll change it :) [07:58:43] I'd use 'rc_cur_id' => $this->mNewRev->getPage() [07:59:03] stuff this, i'll pay an indian freelancer to worry how to do this. [07:59:56] why? [08:00:24] Aloha! [08:00:31] I'm not sure thats indexed [08:00:35] and I don't see where it can be wrong [08:00:44] cur_id is, but timestamp is not in there [08:01:24] can I have co-sysop on my own wiki?! can anybody say me [08:01:33] iobug: sure [08:01:37] just set user rights to someone else too [08:01:44] how? [08:01:49] special:userrights ? [08:02:11] VoiceOfAll: the timestamp is _very_ selective already [08:02:15] sour1, I used xml2sql and it created dumps of 3 tables [08:02:22] 10x, I am not sure, but will try :) [08:02:23] text, revisions and page [08:02:24] domas: page moves don't update rc_ns,rc_title, only rev_page. So getTitle(), getNamespace() point to the new title for the revision [08:02:44] I tried to import them using phpmyadmin [08:02:47] VoiceOfAll: do you mean that page moves update ALL revisions in rev?? [08:03:05] but gives me the following fatal error: Fatal error: Cannot use string offset as an array in E:\Program Files\Apache Software Foundation\Apache2.2\htdocs\phpMyAdmin\libraries\sqlparser.lib.php on line 1014 [08:03:11] ok, you might be right [08:03:22] Should I use it in a special mode? [08:03:52] Hmm, could you potentially save time by taking an sql query, hashing it, and seeing if that hash exists in a memory cache? [08:04:32] domas: it changes page_title,page_ns, but rev_page points to it [08:04:47] yup [08:04:47] oh, sour1 has left the room [08:04:50] you're very very right [08:05:06] domas: but that ns,title,timestamp index would be very good [08:05:21] if only the cur_id index had timestamp [08:06:14] VoiceOfAll: timestamp index is perfect for the job [08:06:19] Does anyone else know the problem of dumps of xml2sql? [08:06:21] hallo, how can i change the time of the wiki? [08:06:27] ns,title,timestamp would help with watchlists though [08:07:23] VoiceOfAll: you know, I can even remove rc_cur_id from conditions [08:07:36] yeah, timestamp may be that good [08:07:49] VoiceOfAll: the very simple reason [08:08:00] VoiceOfAll: entries inside table are ordered by timestamp [08:08:08] so by fetching two subsequent records you always hit same db page [08:08:15] whereas by cur_id it may be scattered [08:08:23] omoo, http://www.mediawiki.org/wiki/Manual:%24wgLocaltimezone [08:09:38] domas: can we add timestamp to the ns,title and cur_id indexes and kill the patrolling index? [08:10:16] I will kill the patrolling index without any changes to indexing [08:10:23] now, adding timestamp to ns,title makes sense anyway :) [08:11:38] wooooo [08:11:39] :) [08:11:42] 03midom * r25525 10/trunk/phase3/includes/DifferenceEngine.php: be resourceful, instead of creating new indexing bloat, use existing indexes. [08:12:02] 03david * r25526 10/branches/liquidthreads/extensions/LqtModel.php: similar caching by id. Threads::where is now only called twice per talk page. [08:12:41] domas: great thanks :) [08:14:31] 03midom * r25527 10/trunk/phase3/maintenance/ (4 files in 3 dirs): destroy rc_patrolling index, no removers added, I hope not that many people did run update with that revision :) [08:14:56] \o/ [08:15:13] always nice to revert something [08:18:20] How can I make sth should be done for table "page" to [08:18:21] hmm [08:18:34] be done with table _Farpage? [08:18:49] what else should I revert? [08:20:36] 03aaron * r25528 10/trunk/phase3/maintenance/updaters.inc: *Remove comment per r25527 [08:20:44] haha, thanks :) [08:21:23] 03kim * r25529 10/trunk/extensions/Wikidata/util/copy.php: [08:21:23] More copy work. Refactor only. [08:21:23] Now overwrite-check ready. [08:21:23] Then relations-ready (although there's still an issue [08:21:23] of uncontrolled recursion) [08:21:24] It runs, so let's ship it. :-P [08:22:03] VoiceOfAll: this is so like me... [08:22:04] http://svn.wikimedia.org/viewvc/mediawiki/trunk/phase3/includes/DifferenceEngine.php?r1=25409&r2=25525 [08:22:17] this is how I explain what I do to Wikipedia to other people [08:22:27] stare at code for few hours, then change single line [08:22:27] :) [08:24:14] good bye (..), [08:25:05] bye! [08:25:09] n8 [08:33:02] domas: your more of a sysadmin and sanity check than a regular dev [08:33:08] *you're [08:33:39] 1 line changes make all the difference ;) [08:34:06] even my mysql patches are oneliners :( [08:41:17] but oh such powerful ones [08:42:59] 03david * r25531 10/branches/liquidthreads/extensions/LqtModel.php: oops [08:53:46] http://en.wikipedia.org/wiki/Russian_Wikipedia#References [08:53:51] ugh, pipe links plz [09:08:05] 03ilabarg1 * r25532 10/branches/ApiEdit_Vodafone/includes/api/ApiEditPage.php: [ApiEditPage.php]-Some comments put by eclipse that produces an error due to a "\n" at the beginning of the class has been cleaned. [09:08:26] Has anyone worked with xml2sql [09:08:27] ? [09:13:28] Has anyone worked with xml2sql? [09:15:30] hi. i'm making a python library that interfaces with mediawiki wikis. e.g. you can load pages with it, save etc - like pywikipedia except instead of focusing on automating tasks it focuses on being a python representation of the interface [09:16:07] i was wondering if someone would like to comment on the syntax i have for it? [09:16:29] ehird, I use pywikipedia [09:17:00] http://mediawiki.pastey.net/73418 [09:17:08] and python has an IDE and I have problems with python IDE for unicode characters [09:17:10] Soroush83: see - "like pywikipedia except instead of focusing on automating tasks it focuses on being a python representation of the interface" [09:17:28] my syntax is quite similar to pywikipedia actually, but much simpler to access arbitary wikis [09:17:29] see my page [09:17:31] *paste [09:18:15] I see and it is the reason I hope it will have less problems with utf-8 if possible. [09:18:32] of course the syntax in my paste is not 100% what it will look like [09:18:34] my problem is mostly because of python IDE [09:18:41] most of the time you will use u"unicode strings" [09:18:44] looking [09:18:44] but its just an example [09:19:50] hmmm. I thought you were talking about making sth like AWB [09:19:58] nope [09:20:00] a python library [09:20:17] that exposes the mediawiki interface in a pythonic, but fully-featured and fully-controllable, way [09:22:31] your code looks like another intrepretion of pywikibot [09:23:42] actullay I still can't see how your code can be more useful than pywikibot. It does what pywikibot [09:23:48] its two examples. [09:23:51] but with different commands [09:23:52] it can do a lot more [09:24:02] what for example [09:24:04] ? [09:24:12] um, i don't see how python can be more useful than machine language [09:24:21] it does what machine language can do but with different commands. [09:24:27] let's all use machine language...? [09:24:49] can it do anything more than what is lyed in api? [09:25:41] I think python is better than c++ or othe programs make machine programs [09:25:54] the one who runs bot must know programming [09:26:18] pywikibots need to be updated all time and always have bugs [09:26:37] They need to have simple code. It can be done in python [09:27:25] you miss my point entirely [09:29:05] I see. but runnig c++ or simillar for a bot looks to be difficult [09:29:35] and the bot drivers need to frequently change the code [09:30:13] It is easier in python [09:31:18] obviously [09:45:18] Hi! [09:46:03] 03aaron * r25533 10/trunk/extensions/ConfirmAccount/ConfirmAccount_body.php: *Treat password as temporary [09:46:18] hi, where_ [09:46:32] can I help you? [09:47:33] yes, writing my questions... [09:48:09] I wonder if there exists a feature, let's call it "zones"/"security levels"? Meaning, 1) one can define that a curtain page/notion belongs to public/private/etc. zone, 2) user/group is defined to be able to access curtain "zones"/"security levels"? [09:48:28] please, help [09:50:22] do you mean $gwGroupPermisions ? [09:50:46] $wqGroupPermisions [09:51:26] I'm not into codding right now, all I'm asking if MediaWiki has this feature inside, available to admin/publisher? [09:52:26] You should change LocalSettings.php anyway [09:52:44] to set your desirable group permissions [09:52:51] which are? [09:54:38] Hi, Malayalam Wiktionary don't have an active community. I would like to improve the project and bring more people. I would like to request for Bureaucrat access. What are the procedures? [09:54:42] ok, I'm reading this on the MediaWiki help... [09:55:10] http://www.mediawiki.org/wiki/Manual:%24wgGroupPermissions [09:55:18] this will help you more [09:56:03] and http://www.mediawiki.org/wiki/Manual:%24wgAddGroups is useful too [09:56:53] ??? [09:57:06] Soroush83: Thanks I will see into it. [09:57:39] you're welcom [09:58:07] brion or another dev [09:58:10] you around? [09:58:16] We have a case of a stolen account [09:58:19] We don't have any developers in our proj [09:59:16] account is [[en:User:Juxo]] [10:00:01] kunsole on irc with a wikimedia cloak is claiming this [10:00:05] hello Sadik [10:00:11] Sadik you have a minute? [10:00:15] hello [10:00:41] hello, why don't you employ some :)) [10:00:44] ? [10:01:46] I am talking about ml.wiktionaty.org :-) [10:02:35] 03mkroetzsch * r25534 10/trunk/extensions/SemanticMediaWiki/specials/URIResolver/SMW_SpecialURIResolver.php: Provide rdf-mime results to RDF-requests [10:02:41] haha [10:02:47] We needs someone with SQL access to enwiki [10:02:58] or ability to set email address some other way [10:02:58] many projects don't have any [10:03:23] :-) [10:04:17] My problem is we don't have an active community. Same here too [10:04:47] Sadik: maybe more complete interface translation could help? [10:05:20] No, [10:05:49] We have to improve, the project, [10:06:16] for that we need bot,sysop... [10:06:28] ...real persons [10:06:57] I can make bot [10:07:24] We need to edit many protected pages [10:08:12] eg: We dont have a page to make new enties [10:08:30] for that we have edit Search page [10:08:47] I think #wikimedia is better for this request ,Sadik [10:09:17] you might want to request for temporary adminship [10:09:50] What about local community approval [10:11:28] you say it doesn't have it [10:11:31] First we have to Improve the Site. then, people will come [10:12:00] I'll try [10:13:26] no response from #wikimedia [10:13:59] Thanks [10:14:03] bye [10:20:53] hey dues [10:21:28] i just read the wikipedia article you left me last night, about patching [10:22:10] i was still confused, but then i read the discussion page on the extension which requires the patch...it turns out i have to install TortoiseSVN to do this [10:26:12] i forgot to remind you i have absolutely no programming exp or anything near it, except html..so yeah i guess this is the 'shell' that was mentioned before [10:30:02] 03kim * r25535 10/trunk/extensions/Wikidata/util/copy.php: (log message trimmed) [10:30:02] Copy safety implemented, same UUID won't be copied twice. [10:30:02] > Recursive copying should now be safe-ish, so can expect new functionalities [10:30:02] wrt recursive copying to appear soon. [10:30:02] !flaws: [10:30:03] * Translated content is treated as a set, so won't do merging [10:30:05] * I'm stealing conceptmapping from the wikidataapi version. [10:42:14] 03(NEW) tog-extendwatchlist wording too cryptic - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=11192 trivial; lowest; MediaWiki: User preferences; (jidanni) [10:44:57] feltipen: you can also use the command line patch utility. you can get it via cygwin, but there's probably also a native windows version aground [10:50:23] 03(NEW) Create "=?UTF-8?Q?=D7=A4=D7=90=D7=A8=D7=98=D7=90=D7=9C?=" namespace on Yiddish Wikipedia - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=11193 15enhancement; normal; Wikimedia: Language setup; (wiki.pedia) [10:52:07] 03(mod) Create "=?UTF-8?Q?=D7=A4=D7=90=D7=A8=D7=98=D7=90=D7=9C?=" namespace on Yiddish Wikipedia - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=11193 (10N/A) [11:30:11] 03kim * r25536 10/trunk/extensions/Wikidata/util/copy.php: [11:30:11] Heh, only one real bug, where I checked for the existence of an [11:30:11] object entry table *before* actually attempting to read it. [11:30:11] Whoops :-P [11:30:11] (runs, basic testing complete) [11:33:06] Hi, is it possible to tweak mediawiki to make it a bit faster? google returns just wikis explaining the word 'performance' :D [11:34:24] q_no: are you using APC/accelerator ? [11:35:44] q_no: http://dammit.lt/2007/01/26/mediawiki-performance-tuning/ [11:35:52] q_no: first hit of 'mediawiki performance tuning' on google. [11:35:53] no. it's a very common apache2 / php 5.22 install on gentoo... no optcode caches [11:36:14] q_no: opcode cache is a must. [11:36:31] indeed [11:37:06] read that page [11:37:08] then come back! [11:37:09] :) [11:37:15] I wish I could switch to lighty/xacache on that machine, but we have to use apache to satisfy our customers :-| [11:37:51]