[00:29:22] <CIA-50>	03aaron * r34606 10/trunk/extensions/FlaggedRevs/ (5 files): 
[00:29:22] <CIA-50>	* Refactor autoreview code
[00:29:22] <CIA-50>	* Improve "0 changes" message
[00:29:22] <CIA-50>	* Check $wgEnableParserCache
[00:29:22] <CIA-50>	* Tweak variable names
[00:29:57] <wikibugs>	03(FIXED) Link changes in flagged pages do not update Special: WhatLinksHere - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14014  +comment (10JSchulz_4587)
[00:30:04] <wikibugs>	03(mod) FlaggedRevs caching/slave lag issues (tracking) - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14029  (10JSchulz_4587)
[00:30:20] <wikibugs>	03(FIXED) Provide other FlaggedRev boxes for the case of a template/ image change, currently resulting in "0 changes need review" - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14031  +comment (10JSchulz_4587)
[00:34:12] <wikibugs>	14(DUP) New parser breaks signatures inside gallery tags - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13120  +comment (10brion)
[00:34:20] <wikibugs>	03(mod) Pre-save transform skips extensions using wikitext (gallery, references, pipe trick, subst, signatures) - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=2700  +comment (10brion)
[00:36:20] <wikibugs>	03(mod) Invalid timestamp in SpecialContributions.php - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14067  (10brion)
[00:37:07] <AaronSchulz>	G_SabinoMullane?
[00:37:30] *AaronSchulz 	points to bug 14067
[00:42:05] <Sundance_Raphael>	hi
[00:42:21] <AaronSchulz>	hey
[00:45:19] <Sundance_Raphael>	moin AaronSchulz 
[00:52:35] <wikibugs>	03(mod) RSS feed for Special:Contributions/newbies - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=6955  (10Wiki.Melancholie)
[00:55:15] <wikibugs>	03(NEW) Short notation links to subpages don' t work in edit summaries - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14080 minor; normal; MediaWiki: Page editing; (wikimedia)
[00:56:00] <Schroeder2>	if I have upload_by_url permission, how do I actually do that?
[00:56:08] <Skizzerz>	Schroeder2: visit Special:Upload
[00:56:16] <Schroeder2>	yeah
[00:56:18] <Schroeder2>	I'm there
[00:56:21] <Schroeder2>	there's nothing different
[00:56:25] <Skizzerz>	hmm
[00:56:41] <Skizzerz>	set $wgAllowCopyUploads = true; in LocalSettings.php
[00:56:47] <Skizzerz>	!copyuploads
[00:56:47] --mwbot--	I don't know anything about "copyuploads".
[00:56:52] <Skizzerz>	silly bot
[00:57:22] <Skizzerz>	http://www.mediawiki.org/wiki/Manual:Configuring_file_uploads#Uploading_directly_from_a_URL for more info
[00:58:01] <Schroeder2>	ok, that worked
[00:58:02] <Schroeder2>	thanks
[00:58:03] <wikibugs>	03(mod) Show TOC on Special:SpecialPages - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13862  +comment (10innocentkiller)
[01:29:00] <Schroeder2>	I have MW configured to allow me to import directly from Wikipedia with $wgImportSources[] = "Wikipedia";
[01:29:11] <Schroeder2>	now how would I go about importing something not in the main namespace?
[01:29:20] <Schroeder2>	for example, if I put "Template:cleanup" in the box
[01:29:45] <shaiaqua>	In what file are the tabs (edit, history, discuss, etc) located?
[01:29:52] <Schroeder2>	I get "Import failed: Couldn't open import file"
[01:30:35] <Schroeder2>	shaiaqua: you mean their text contents?
[01:30:41] <Schroeder2>	they're in various system messages
[01:30:46] <Schroeder2>	Special:Allmessages will give you a list
[01:30:58] <shaiaqua>	no, the physical tab
[01:31:02] <shaiaqua>	I want to delete one
[01:32:55] <Suisuisui>	from all of your wiki?
[01:33:26] <Suisuisui>	which do you like xhtml+PHP or css 
[01:33:41] <shaiaqua>	either or
[01:34:10] <Suisuisui>	hmm :(
[01:34:21] <shaiaqua>	?
[01:34:54] <Suisuisui>	skin/monobook.php will show you the skin code
[01:35:12] <Suisuisui>	or you can edit Mediawiki:common.css to delete it 
[01:35:39] <Suisuisui>	or you can delete it with your user style sheet on User:foobar/Monobook.css
[01:36:04] <shaiaqua>	and to add a tab?
[01:36:10] <Suisuisui>	same
[01:36:48] <Suisuisui>	to add is litle difficult than remove
[01:37:22] <shaiaqua>	common.css is empty?
[01:37:28] <Suisuisui>	you need to add your tab's to main.css
[01:37:35] <Suisuisui>	yes in default
[01:38:33] <Suisuisui>	see http://en.wikipedia.org/wiki/Mediawiki:common.css or somewhere you like 
[01:38:42] <shaiaqua>	I don't see how the tabs can be written entirely in css
[01:38:51] <Suisuisui>	:|
[01:39:23] <Suisuisui>	read skin/monobook/main.css 
[01:42:43] *shaiaqua 	can't find more than margins and padding for the tabs
[01:42:55] <flyingparchment>	the key part is li { display: inline; }
[01:43:02] <flyingparchment>	that makes them go next to each other instead of in a vertical list
[01:43:57] *shaiaqua 	is looking for where the tabs are written
[01:46:58] <shaiaqua>	Suisuisui: I can't find where the tabs are written in mais.css
[01:47:30] <Suisuisui>	ooh
[01:48:12] <shaiaqua>	main*
[01:48:14] <Skizzerz>	shaiaqua: should be #p-cactions
[01:48:40] <Suisuisui>	yup
[01:49:25] <Suisuisui>	there's not written a tab directry , writen with XHTML on skin/monobook.php
[01:49:26] <wikibugs>	03(NEW) Show section edit links on stable versions when there is a more recent draft - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14081 15enhancement; normal; MediaWiki extensions: FlaggedRevs; (gmaxwell)
[01:50:08] <Suisuisui>	so you need to add a tab with dom and some js
[01:50:15] <Suisuisui>	or .. php
[01:50:18] <shaiaqua>	so changes have to be made to all skins....
[01:50:43] <Suisuisui>	yes
[01:51:08] <Suisuisui>	(as I think)
[01:51:20] <Arana1>	adding tabs is done from the SkinTemplateTabs hook
[01:52:36] <Suisuisui>	making new extension :D
[01:53:49] <shaiaqua>	:)
[01:55:00] *Soxred93 	thinks his new extension is very useful, and should be installed on Wikimedia, but will not due to it being in beta and that he has a COI with it.
[01:55:36] <Suisuisui>	:|
[01:57:55] <shaiaqua>	that took me a while :|
[01:58:56] <Suisuisui>	hmm
[01:59:40] <Soxred93>	My new extensions, not Suisuisui's.
[02:00:24] <Suisuisui>	what
[02:00:26] <Suisuisui>	?
[02:00:26] <Arana1>	http://www.organicdesign.co.nz/MediaWiki_code_snippets#Adding_a_new_action_tab
[02:00:59] <Soxred93>	Which I just realized means "is uisui"
[02:01:15] <Soxred93>	never mind
[02:08:01] <CIA-50>	03dale * r34607 10/trunk/extensions/MetavidWiki/skins/mv_embed/ (3 files): 
[02:08:01] <CIA-50>	switched over to cortado r34333 from wikimedia
[02:08:01] <CIA-50>	(includes a few bug fixes see oggHandler for details)
[02:17:18] *AaronSchulz 	hopes G_SabinoMullane can look at bug 14067
[02:20:31] <CIA-50>	03nad * r34608 10/trunk/phase3/maintenance/mssql/README: add some dev notes and other info
[02:22:16] <wikibugs>	03(NEW) Ajax search bug with complex IMEs - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=14082 15enhancement; normal; MediaWiki: Search; (kjoonlee)
[02:22:27] <Werdna>	hi
[02:24:08] <wikibugs>	03(mod)  Enable editors to control footnote ordering and to insert wikitex between footnotes . Also declutter article prose. - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=12796  (10maury.markowitz)
[02:53:30] <shaiaqua>	if I want to require /home/shaiaqua/public_html/w/extensions/Renameuser/SpecialRenameuser.php from /home/shaiaqua/public_html/w/LocalSettings.php, what do I put in the require? Everything I've tried so far has failed
[02:54:39] <AaronSchulz>	shaiaqua: are /extensions  and localsettings.php in the same folder?
[02:54:41] <AphelionZ>	Hey everybody, I'm having some API performance issues. Is anybody from that crew around?
[02:54:55] <shaiaqua>	yea
[02:55:03] <AaronSchulz>	AphelionZ: heh, thats API ;)
[02:55:16] <AphelionZ>	o rly
[02:55:21] <AphelionZ>	is that a known issue?
[02:55:27] <AphelionZ>	that it's slow as heck
[02:55:35] <AaronSchulz>	shaiaqua: and is your require variable the actual path, not the url?
[02:56:12] *AaronSchulz 	has include_once('extensions/Renameuser/SpecialRenameuser.php');
[02:56:17] <shaiaqua>	yea, I've tries $IP/extensions..., extensions..., /extensions... and all have failed
[02:57:08] <AaronSchulz>	AphelionZ: WMF api?
[02:57:11] <AaronSchulz>	what query?
[02:57:20] <shaiaqua>	still Warning: require_once(extensions/Renameuser/SpecialRenameuser.php) [function.require-once]: failed to open stream: No such file or directory in /home/shaiaqua/public_html/w/LocalSettings.php on line 147
[02:58:05] <AphelionZ>	WMF = Wiki Media Foundation?
[02:58:14] <AphelionZ>	I'm talking about the one that comes packaged with 1.12..
[02:58:23] <AphelionZ>	and also the Vodafone branch
[02:58:47] <AaronSchulz>	shaiaqua: is actually there ;) ?
[02:58:55] <shaiaqua>	yea :p
[02:59:09] <AaronSchulz>	what is the directoy?
[03:00:39] <shaiaqua>	 /public_html/w/extensions/RenameUser
[03:01:00] <AaronSchulz>	and localsettings.php is in /w/ ?
[03:01:29] <AaronSchulz>	shaiaqua: are you on linux?
[03:01:40] <shaiaqua>	yea
[03:01:51] <shaiaqua>	it is and I am
[03:01:56] <AaronSchulz>	include_once('extensions/RenameUser/SpecialRenameuser.php');
[03:01:58] <AaronSchulz>	do that
[03:02:45] <AaronSchulz>	AphelionZ: what site is this?
[03:03:03] *AaronSchulz 	isn't sure if that branch is stable :p
[03:03:26] <shaiaqua>	still no
[03:03:46] <AphelionZ>	working on http://wikibondev.org/wiki/v/Home
[03:04:07] <AaronSchulz>	shaiaqua: so "/public_html/w/extensions/RenameUser" is where it is right?
[03:04:23] <shaiaqua>	yup
[03:04:25] <AaronSchulz>	the case must be right
[03:05:02] <shaiaqua>	it is
[03:05:17] <AaronSchulz>	do you have other extensions?
[03:05:52] <shaiaqua>	yes
[03:06:02] <shaiaqua>	oddly that is the only one complaining
[03:07:22] <AphelionZ>	AaronSchulz: here's a log file of API calls I use to load a single 'wiki' page on my site
[03:07:24] <AphelionZ>	http://pastebin.com/d13038fe7
[03:08:21] <AaronSchulz>	shaiaqua: what do those include statements look like?
[03:08:25] *shaiaqua 	is confused
[03:08:32] <shaiaqua>	you want pastebin?
[03:08:35] <AaronSchulz>	yeah
[03:09:08] <AaronSchulz>	AphelionZ: do you have profiling on?
[03:09:29] <AphelionZ>	yeah, you can see it at the bottom of the page: http://wikibondev.org/wiki/v/The_most_telling_part_of_BT%27s_virtualization_strategy
[03:09:57] <shaiaqua>	it works now
[03:10:06] <AaronSchulz>	;)
[03:10:16] <AaronSchulz>	shaiaqua: what did you change?
[03:10:43] <AaronSchulz>	shaiaqua: note that if you get fatal error, you often need to purge the browser cache to retry the page
[03:10:57] <shaiaqua>	that might be it
[03:11:47] <AaronSchulz>	AphelionZ: do you have one unified profiling page?
[03:12:04] <AphelionZ>	no, no i dont
[03:12:08] <AphelionZ>	i dont even really know what you mean
[03:12:12] <AaronSchulz>	hehe
[03:12:20] <AaronSchulz>	AphelionZ: you can use startprofiler
[03:12:50] <AphelionZ>	ohh the MW thing... does that work with the API?
[03:13:04] <AphelionZ>	i dont have access to the functions since im going at them through CodeIgniter
[03:13:05] <AaronSchulz>	should, API, uses same DB class
[03:13:23] <AphelionZ>	i'd have to edit the API code
[03:13:27] <AphelionZ>	what are you trying to diagnose
[03:13:29] <AaronSchulz>	why?
[03:13:44] <AaronSchulz>	the API code already makes profile calls
[03:13:51] <AaronSchulz>	you just have to turn on profiling
[03:15:15] <AaronSchulz>	AphelionZ: anyway, slow queries should show up
[03:15:22] <AphelionZ>	ok, like in LocalSettings.php
[03:15:29] <AphelionZ>	i'll look up profiling in the documentation
[03:15:37] <AaronSchulz>	AphelionZ: the documentation is lame
[03:15:49] <AphelionZ>	then tell me exactly what to do :D :D
[03:15:51] <AaronSchulz>	AphelionZ: just go to StartProfiler.php
[03:15:56] <AaronSchulz>	do what it says there
[03:16:00] <AphelionZ>	k sweet
[03:16:27] <AaronSchulz>	then, you will want to SVN up probably
[03:16:49] <AaronSchulz>	some of the profileinfo stuff was broken, so I had to fix it a while ago
[03:17:27] <AphelionZ>	my version right now is 1.13alpha (r33497)
[03:17:29] <AaronSchulz>	http://svn.wikimedia.org/viewvc/mediawiki/trunk/phase3/maintenance/archives/patch-profiling.sql?revision=33997&view=markup&sortby=date
[03:17:34] <AaronSchulz>	then, run that query
[03:17:34] <AphelionZ>	where does the profile output go
[03:17:54] <shaiaqua>	where can I change to make tabs to empty discussion pages not add &action=edit?
[03:18:05] <AaronSchulz>	you can make it go to a lot of places, but the above query is for a HEAP table
[03:18:30] <AaronSchulz>	AphelionZ: make sure you replace /*$wgDBprefix*/ with the right thing, like mw_ or whatever
[03:19:29] <AaronSchulz>	AphelionZ: yeah, SVN up
[03:19:57] <AphelionZ>	ugh I can't right now, no SSH access from my IP
[03:21:10] <AaronSchulz>	AphelionZ: so after editing startprofiler, running that query, SVN upping, then edit localsettings.php
[03:21:13] <AaronSchulz>	add $wgProfileToDatabase = true;
[03:21:54] <AaronSchulz>	AphelionZ: when you say API, do you mean api.php?
[03:21:59] <AaronSchulz>	or just general MW stuff?
[03:22:07] <AphelionZ>	well look at my logs
[03:22:13] <AphelionZ>	some of them im calling action=raw from index.php
[03:22:19] <AphelionZ>	and some of them im calling api.php
[03:22:24] <AphelionZ>	but I'm pretty sure I mean api.php
[03:22:53] <AaronSchulz>	well, profiling covers it all
[03:23:00] <AaronSchulz>	so when you get it up, it should show
[03:23:05] <AphelionZ>	ok cool
[03:23:12] <AphelionZ>	so yeah i just need to svn
[03:23:14] <AphelionZ>	i did everytihng else
[03:23:27] <AphelionZ>	also is that revision number for vodafone or for trunk
[03:23:58] <AaronSchulz>	oh, yeah, is adminsettings.php set?
[03:24:18] <flyingparchment>	profileinfo.php!
[03:24:33] <AphelionZ>	naw i just got the sample
[03:25:11] <AaronSchulz>	the sample should say what to do
[03:27:03] <AphelionZ>	alright, did that
[03:27:09] <AaronSchulz>	AphelionZ: after that, add $wgDBprefix to adminsettings.php
[03:27:10] <AphelionZ>	profile table is already flooding with data :D
[03:27:14] <AaronSchulz>	then set $wgEnableProfileInfo = true;
[03:27:49] <AaronSchulz>	AphelionZ: do you have a prefix?
[03:27:59] <AphelionZ>	yeah its mediawiki_
[03:28:17] <AphelionZ>	153 rows in profiling table? sound about right?
[03:28:32] <AaronSchulz>	if you don't feel like waiting to SVN up, you can make the 2 line changes to profileinfo now
[03:28:41] <AphelionZ>	ok i'll do that
[03:29:39] <Fire>	Hi.
[03:31:17] <Charitwo>	hello
[03:31:23] <AphelionZ>	Hi! :D
[03:31:31] <AphelionZ>	hiya, even
[03:35:31] <AphelionZ>	ok so If i run profileinfo.php, and filter by API I get: http://pastebin.com/d324dfd40
[03:35:37] <AaronSchulz>	 	 Profiling error:
[03:35:44] <AphelionZ>	what does it mean? the API is just slow as hell?
[03:35:44] <AaronSchulz>	indicates unmatched in/out calls
[03:36:05] <AaronSchulz>	that is the fault of some API dev ;)
[03:36:18] <AphelionZ>	interesting
[03:36:26] <AphelionZ>	doesnt look like it takes up any time though?
[03:38:27] <AaronSchulz>	sec
[03:39:37] <simprix>	Im trying to use the LDAP Authentication plugin. My active directory is called domain1.local. What line to I specify the ip address of the Primary Domain Controller. 
[03:40:32] <AaronSchulz>	AphelionZ: so, is the problem just some slow queries, or are page views to slow?
[03:40:47] <AphelionZ>	the queries dont appear to be slow: http://wikibondev.org/w/profileinfo.php?filter=API&sort=time&expand=
[03:41:33] <AphelionZ>	but the page loads are just sooo sluggish and the codeigniter profiling tells me its taking upwards of 5-6 seconds per load, which, combined with upload/download speeds on peoples connections comes to a good 10-20 second page load
[03:41:50] <AaronSchulz>	AphelionZ: do you have APC or eaccelerator ?
[03:42:07] <AphelionZ>	no..?
[03:42:12] <AaronSchulz>	eeks
[03:42:22] <AaronSchulz>	you will want to
[03:43:35] <AphelionZ>	really? the solution to this is to add yet another piece of software to the puzzle?
[03:43:54] <AaronSchulz>	http://dammit.lt/2007/01/26/mediawiki-performance-tuning/
[03:44:02] <AaronSchulz>	AphelionZ: PHP is kinda lame is some ways
[03:44:11] <AaronSchulz>	it through the opcode away each hit
[03:44:21] <AaronSchulz>	and redoes it next time
[03:44:39] <AaronSchulz>	AphelionZ: eaccelerator is easy to install
[03:44:52] <AaronSchulz>	haven't fiddled with APC, but thats what wikipedia uses :)
[03:44:57] <AphelionZ>	holy crap 217 down to 37
[03:45:08] <AphelionZ>	so APC + eaccelerator eh
[03:45:18] <shaiaqua>	my logo disappears at times...  
[03:45:35] <AaronSchulz>	AphelionZ: well, one or the other :)
[03:45:41] <AphelionZ>	oh ok haha
[03:45:49] <AaronSchulz>	CACHE_ACCEL is just a shitty constant name :D
[03:46:02] <wikibugs>	03(NEW) "The following 200 pages are in this category" - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14083 15enhancement; normal; MediaWiki: Categories; (jidanni)
[03:46:28] <AaronSchulz>	AphelionZ: also, set $wgMiserMode in localsettings.php
[03:46:40] <AaronSchulz>	if you really want to avoid anything slow
[03:47:11] <AphelionZ>	what does that do
[03:47:30] <AaronSchulz>	various slow/cruft features are turned off
[03:47:33] <AaronSchulz>	wikipedia has it on
[03:47:43] <AaronSchulz>	you don't really lose much
[03:48:32] <AphelionZ>	yeah
[03:48:39] <AphelionZ>	was gonna say, not a notable difference
[03:48:44] <AphelionZ>	but every little bit helps
[03:48:53] <AphelionZ>	this APC thing is gonna be cool I think
[03:48:59] <AaronSchulz>	well, wikipedia would overload if it was off ;)
[03:49:17] <AaronSchulz>	various filesorts and what not would melt our servers :(
[03:49:23] <AphelionZ>	hah
[03:49:48] <AphelionZ>	tbh i can live with a sluggish site, but once the timeouts start happening then my clients start calling and whining
[03:50:08] <shaiaqua>	my logo disappears if I don't also require it, any ideas why?
[03:50:35] <AphelionZ>	what about caching? does MW have anything like that?
[03:50:45] <shaiaqua>	memcaching?
[03:50:46] <AphelionZ>	and if so does the API benefit from it 
[03:51:03] <AaronSchulz>	AphelionZ: not much api.php caching
[03:51:08] <AaronSchulz>	but pages can be cached
[03:51:17] <AaronSchulz>	the HTML that is
[03:51:21] <AphelionZ>	well im rarely viewing wiki pages
[03:51:31] <AaronSchulz>	your viewers are :)
[03:51:36] <AaronSchulz>	HTML cache helps a lot
[03:51:40] <AphelionZ>	bring it on!
[03:51:43] <AaronSchulz>	all of which can leverage off APC
[03:52:14] <AphelionZ>	but what i mean to say is that wikibon will be viewing all of the wiki content through the API
[03:52:23] <AaronSchulz>	ahh
[03:52:34] <AaronSchulz>	well, it really just depends on the type of query
[03:52:44] <AphelionZ>	we're going to leave access to the "vault" which will be the raw mediawiki install
[03:52:52] <AaronSchulz>	most of API is fast, just some wierd queries/combinations that get really slow
[03:52:59] <AphelionZ>	but for the most part it'll be the /wiki/v/ codeigniter pages
[03:53:16] <AaronSchulz>	AphelionZ: still, APC lets you use revision cache and diff cache
[03:53:30] <AphelionZ>	http://pastebin.com/d13038fe7 <-- yeah those are the main queries i'll be doing in loading a typical wiki article
[03:53:31] <AaronSchulz>	useful if you have external storage or compress
[03:53:35] <AaronSchulz>	*compression
[03:53:46] <AaronSchulz>	or even if not, still better than DB hits
[03:53:49] <AphelionZ>	sounds great.. dont have either of those yet but i'm sure we will
[03:55:29] <AphelionZ>	alright so the next step is that i should be looking at APC huh
[03:55:38] <AaronSchulz>	right
[03:55:55] <AphelionZ>	ok great
[03:56:00] <AphelionZ>	thank you so much Aaron
[03:56:15] <wikibugs>	03(NEW) FSRepo does not work well - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14084 normal; normal; MediaWiki: Images; (mediawiki-bug)
[03:56:32] <AaronSchulz>	AphelionZ: np :)
[03:57:50] <wikibugs>	03(mod) FSRepo does not work well - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14084  +comment (10JSchulz_4587)
[04:00:11] <barosl>	wow, you guys are really fast. i submitted a bug report, and there was a response within 1 minute.
[04:01:05] <Charitwo>	thats what the bot is for :)
[04:01:09] <flyingparchment>	that's because we're all losers who sit on irc all day and watch the bot that posts requests here
[04:01:35] <Charitwo>	oh river
[04:02:23] <barosl>	lol
[04:03:54] <Splarka>	fp: WP:BEANS!
[04:04:12] <flyingparchment>	no, i'm pretty sure everyone knows that ;)
[04:04:54] <Charitwo>	Splarka: baked or refried?
[04:11:35] <shaiaqua>	my logo disappears if I don't also require it, any ideas why?
[04:14:04] <shaiaqua>	is it possible to make broken links to talk/discussion pages not lead to the &action=edit?
[04:14:54] *AaronSchulz 	was just talking to some guys
[04:16:05] *AaronSchulz 	hugs flyingparchment
[04:19:08] *shaiaqua 	waits patiently
[04:19:48] <Nikerabbit>	hello
[04:19:55] <shaiaqua>	hi
[04:20:16] <AaronSchulz>	alright, time for a drink
[04:20:57] *shaiaqua 	continues waiting patiently
[04:24:42] <flyingparchment>	you don't sound very patient to me
[04:25:32] <shaiaqua>	:p
[04:27:13] <Splarka>	function makeBrokenLinkObj() in Linker.php handles it
[04:28:00] <Splarka>	looks rather hard coded eh
[04:28:23] <shaiaqua>	can I make that call makeKNownLinkObj()?
[04:30:44] <shaiaqua>	t broke
[04:30:47] <shaiaqua>	:(
[04:32:10] <Crazy_Louco>	4[Lestaty4]: you is liar
[04:32:29] <Lestaty>	:S lol
[04:32:50] <Crazy_Louco>	4[Lestaty4]: cade o cara?
[04:33:11] <Lestaty>	esquece... eu nao vou falar :P
[04:33:35] <Crazy_Louco>	12[Lestaty12]1: seu egoista!
[04:33:57] <Lestaty>	nao é egoismo é senso de educação :P
[04:34:05] <Crazy_Louco>	uia
[04:35:15] <shaiaqua>	didn't work :(
[04:36:05] <Splarka>	shaiaqua: try changing "$q = 'action=edit&redlink=1';" to "$q = 'action=view';"  maybe
[04:38:33] <shaiaqua>	still no
[04:46:38] <shaiaqua>	*cry*
[04:47:57] *Splarka 	shrugs
[04:57:24] <KTC>	one can do $wgUser->isBlocked() to find out if a user is blocked, is there similar for checking if the user have the bot flag?
[04:58:22] <Arana1>	isBot()
[04:59:20] <KTC>	doh. i searched for "bot " in User.php. thanks :)
[06:12:21] <CIA-50>	03nad * r34609 10/trunk/phase3/maintenance/sqlite/tables.sql: replace with newly generated version
[06:13:24] <bit`>	how can I make [[Image]] link to an article instead of the full sized picture
[06:13:44] <Splarka>	!linkedimage
[06:13:44] --mwbot--	I don't know anything about "linkedimage".
[06:13:49] <Splarka>	!imagelink
[06:13:49] --mwbot--	Image linking is intentionally not supported in MediaWiki, so access to the image's description page is always available. If you need image links, see <http://www.mediawiki.org/wiki/Help:Linked_images> for methods and information.
[06:13:56] *Splarka 	brainfarts
[06:14:53] <Werdna>	eh?
[06:15:02] <Werdna>	bit`: use [[:Image:someimage.jpg]]
[06:15:28] <bit`>	hm
[06:15:34] <bit`>	:Image?
[06:16:03] <Werdna>	yes
[06:16:06] <Arana1>	that would make the image render as a normal link
[06:16:26] <Arana1>	similar to linking to a category page
[06:16:54] <wikibugs>	03(NEW) Please switch to Wiki.png at lo.wiktionary.org - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14085 15enhancement; normal; Wikimedia: Bugzilla; (vienglao)
[06:17:32] <bit`>	makes the image disappear
[06:17:51] <bit`>	guess there's no such template for this wiki
[06:18:11] <Arana1>	leading colon just means to render as a normal link
[06:18:12] <Splarka>	[[:Image:Whatever]] is how you wikilink to a full sized image
[06:18:24] <Splarka>	what you want is a linked image, see the help link above
[06:18:34] <Splarka>	that one: http://www.mediawiki.org/wiki/Help:Linked_images
[06:19:35] <bit`>	thanks
[06:30:57] <wikibugs>	03(mod) Increase rate limit for rollbacker group to 10 per minute - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=12760  +comment (10FastLizard4)
[07:00:18] <_wooz>	lo
[07:00:37] <Charitwo>	hello
[07:32:49] <e-BIT>	yarr
[07:32:59] <wikibugs>	03(mod) Gender switch in user preferences - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=13040  (10xarak)
[07:33:22] <Charitwo>	wb bit^ 
[07:33:33] <bit^>	all the nicks are taken on this server!
[07:33:46] <bit->	mah god
[07:34:02] <bit^^>	ok..
[07:34:51] <Charitwo>	bit^^: bit^ is available to be dropped, you can request it in #freenode
[07:35:07] <Charitwo>	just /msg nickserv info bit^
[07:35:14] <Charitwo>	anything over 60 days can be dropped
[07:36:24] <bit^^>	thanks
[07:36:26] <bit^^>	so... is there any simple way to scale images like [[Image]] does
[07:36:50] <Charitwo>	you can do a pixel value
[07:37:00] <bit^^>	that means?
[07:37:00] <Charitwo>	[[Image:Example.jpg|250px]]
[07:37:15] <bit^^>	yeah, I mean without [[Image]]
[07:37:41] <bit^^>	http://cookiepedia.nfshost.com/index.php/Template:ImageLink
[07:38:04] <bit^^>	I need to get the height value somehow
[07:38:17] <Charitwo>	not sure, my brain isn't the best at 4am
[07:38:28] <bit^^>	heh
[07:44:03] <dvdb>	hi
[07:44:26] <Charitwo>	hello
[07:46:45] <Fire>	dvb: What about asking your question :)
[07:47:04] <dvdb>	a question to table "flaggedpages" (flagged extension; used in dewiki): the version 44394454 of article "1397" was marked as "reviewed", the same version was two minutes unflagged (as unreviewed/unsighted). but the query "SELECT fp_reviewed, fp_stable FROM `flaggedpages` WHERE `fp_page_id` = 7287;" returns stil 1 and 44394454.
[07:48:37] <dvdb>	how do I know whether the newest version of an article is stil marked as reviewed? - fp_reviewed should be 0, but it isnt :/
[08:47:16] <wikibugs>	03(mod) Invalid timestamp in SpecialContributions.php - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14067  (10N/A)
[08:56:06] <wikibugs>	03(NEW) Allow filtering user contributions by flags - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14086 15enhancement; normal; MediaWiki: Special pages; (chenzhengwei2002)
[09:02:28] <wikibugs>	03(mod) Please switch to Wiki.png at lo.wiktionary.org - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14085   +shell (10raimond.spekking)
[09:03:04] <wikibugs>	03(mod) Configure groups in nowiki for access to Special:Unwatchedpages - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14079   +shell (10raimond.spekking)
[09:05:19] <NicDumZ>	erm. Trying to put a page with *a lot* of page inclusions has strange consequences, and the end-user interface is *not* good.
[09:06:28] <NicDumZ>	1) if you put too many page inclusions in a page, but if you keep the number reasonable, the last page inclusions won't be included, but will be shown as blue links. Why not, but no notice is being sent to the user
[09:12:36] <Splarka>	view the page source
[09:13:49] <Splarka>	look for a comment indicating the transclusion/expensive parser function/node limits
[09:15:35] <NicDumZ>	2) If you *really* try to put too many templates in the page, you'll get one of these http errors : 1) 503 (blue html error message, ERR_READ_TIMEOUT) 2) 500 3) 200 (White plain page)
[09:16:25] <NicDumZ>	mm yes Splarka that's what I found out searching in bugzilla...
[09:19:32] <Splarka>	the limits are there to try to prevent problems like that ^_^
[09:19:44] <Splarka>	the timeout problem can come from too many links or too much text too
[09:20:10] <Splarka>	"no system is foolproof because fools are so ingenious"
[09:24:39] <NicDumZ>	(oh and the problem was "WARNING: template omitted, post-expand include size too large". )
[09:37:25] <Nikerabbit>	hello
[09:37:43] <Fire>	Hi
[09:38:29] <Dantman>	NicDumZ: That's something that will never be fixed
[09:38:41] <Dantman>	That's an issue with the Old parser which has since been replaced
[09:40:54] <wikibugs>	03(mod) Install ConfirmAccount extension on enwiki - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13782  +comment (10cometstyles)
[09:41:52] <NicDumZ>	Dantman, and... the new parser is not yet on WM sites, right ?
[09:44:00] <Dantman>	Nope, it's running
[09:44:55] <Dantman>	The old Parser is depreciated /because/ WikiMedia is using it already
[09:45:20] <NicDumZ>	then I don't get you Dantman, that error is from fr.wp, and you tell me it came from the old parser ?! :)
[09:45:37] <Dantman>	bleh...
[09:45:46] <Dantman>	Gimme a page?
[09:46:07] <NicDumZ>	Dantman, http://fr.wikipedia.org/wiki/Utilisateur:NicDumZ/Test :)
[09:47:31] <Werdna>	Post-expand include size: 2048000/2048000 bytes
[09:47:42] <NicDumZ>	yes :)
[09:47:57] <Dantman>	ok, new parser
[09:48:04] <NicDumZ>	ahah :þ
[09:48:32] <Dantman>	But honnestly
[09:48:39] <Dantman>	That page there is insane
[09:49:02] <NicDumZ>	It is meant to be a Project:Translation watchlist page. It works for every languages but en: :þ
[09:49:29] <Dantman>	If it takes me over a minute to list out the source text of the page... It's to long
[09:49:36] <Dantman>	Much less actually showing the page
[09:50:03] <Dantman>	That page is so long, that people aren't even going to be able to read it.
[09:50:27] <Dantman>	If the PP doesn't break untill you add in that much data, I don't think it's broken at all
[09:50:33] <NicDumZ>	Dantman, http://fr.wikipedia.org/wiki/Projet:Traduction/*/Lang/it for example is nice. It's just not scalable for en: translations :)
[09:52:45] <Dantman>	The sollution here is not to change the parser... It's to be sane and split your data among multiple pages
[09:52:53] <Splarka>	indeedy
[09:52:57] <Splarka>	"foolproof"
[09:53:04] <Jack_Phoenix>	hey Dantman and Splarka :)
[09:53:14] *Splarka 	grrars at jack
[09:54:00] <Dantman>	Honnestly... The PP should probably cut out even sooner... To prevent output of pages with an insane ammount of input
[09:54:12] <NicDumZ>	oh, yes, and that's what I concluded by myself. I was not expecting anyone to change the parser for a this-extreme case :)
[09:54:54] <Dantman>	I honnestly think that the current behaviour for when the parser can't transclude is good 
[09:55:21] <Dantman>	The reader is not meant to be given an error when the editor makes a bad move.
[10:11:22] <NicDumZ>	Dantman, no, the reader should not get an error. But the *editor* does not get an error. And it should perhaps warn him explicitely. Checking for the HTML source is not really user friendly ;)
[10:11:54] <Dantman>	Yup... but that's something outside the scope of the parser
[10:12:10] <Dantman>	^_^ You're asking for a UI enhancement
[10:12:31] <Dantman>	Though, there is a potential issue with that
[10:12:59] <Dantman>	Wiki do migrate parsers, so there is potential for a wiki to switch and these issues go unoticed
[10:13:14] <NicDumZ>	Oh yes I should have made it clear, I was not whining about the parser behavior, just that the UI on parser limitations was not this good :)
[10:14:03] <Splarka>	easy to make a node parser JS that alerts you to any interesting comments...
[10:14:20] <Splarka>	might be a handy greasemonkey or gadget for mediawiki
[10:15:20] <Splarka>	just iterate recursively over the node tree and if .nodeType == 8, it is a comment, grab and parse the .nodeValue
[10:17:38] <Dantman>	Plug that into Popups maybe?
[10:26:36] <Splarka>	it'd be slow, heh
[10:28:40] <Splarka>	ahh, here it is: http://en.wikipedia.org/wiki/User:Splarka/showcomments.js
[10:29:01] <Splarka>	only checks for .search(/(NewPP|Served by)/) as is
[10:29:33] <Splarka>	could do it on all page loads looking for errors
[11:20:21] <Nikerabbit>	      6 PHP Notice:  Use of undefined constant FOR_UPDATE - assumed 'FOR_UPDATE' in /var/www/w/includes/Title.php on line 200
[11:21:39] <Nikerabbit>	aarons code, puh
[11:21:44] <domas>	heh
[11:25:11] <Nikerabbit>	is it supposed to be GAID_FOR_UPDATE or what?
[11:26:04] <Nikerabbit>	./extensions/FlaggedRevs/FlaggedRevs.php:               $title = $rev->getTitle() ? $rev->getTitle() : Title::newFromID( $rev->getPage(), FOR_UPDATE );
[11:26:07] <Nikerabbit>	another instance
[11:27:28] <CIA-50>	03nad * r34610 10/trunk/phase3/includes/DatabaseMssql.php: pages can render, but many operations break it
[11:38:26] <MONK3000>	Hi. I want the User Tool Links to link to a Special:PM/[User] page instead of the Talk page. i think it's quite easy and i'm too dumb. atm it reads: $userTalkPage = Title::makeTitle( NS_USER_TALK, $userText );
[11:39:03] <MONK3000>	so in my logic it would replace NS_USER_TALK by 'Special:Pm/' but that doesn't owrk of course
[11:43:05] <MONK3000>	ok got it. no worries
[11:49:06] <MONK3000>	next problem: What is the best way to disable a namespace? For example I don't need User Talk. So is it better to restrict permissions or "$wgNamespaceAliases - override the default names for namespaces."
[12:07:02] <shaiaqua>	how can links to nonexistent talk/discussion pages be made not to go to action=edit?
[12:19:45] <Werdna>	shaiaqua: hax
[12:20:29] <shaiaqua>	to what? I can't find where that is writen
[12:20:58] <Werdna>	uh, grep source for redlink=1
[12:21:03] <Werdna>	or something like that
[12:23:40] <wikibugs>	03(mod) Allow filtering user contributions by flags - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14086  +comment (10dungodung)
[12:23:46] <dungodung>	\0/
[12:34:55] <wikibugs>	03(mod) Allow filtering user contributions by flags - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14086  (10N/A)
[12:50:07] <shaiaqua>	nothing came up
[12:53:30] <MONK3000>	hmh i would liek to knwo that as well and would like to disable teh red user links. besides the fact that i use the extension "Keep you hands to yourself" i.e. you cannot edit other people's user pages. so every user link should be blue leading to the user page (existent or not)
[13:05:12] <Nikerabbit>	AaronSchulz: hello
[13:05:19] <AaronSchulz>	heya
[13:05:23] <MGrab>	hello all
[13:05:24] <Nikerabbit>	6 PHP Notice:  Use of undefined constant FOR_UPDATE - assumed 'FOR_UPDATE' in /var/www/w/includes/Title.php on line 200
[13:05:40] <Nikerabbit>	one in core and one in flaggedrevs
[13:06:07] *AaronSchulz 	isn't getting that. I'll try again
[13:06:28] <Nikerabbit>	is it supposed to be GAID_FOR_UPDATE?
[13:07:07] <AaronSchulz>	no
[13:07:10] <AaronSchulz>	I'll just add define ( 'FOR_UPDATE', 1 );
[13:07:11] <Dantman>	T_T I really need to find a way to notify myself when it's becoming late
[13:07:13] <shaiaqua>	can all broken links be made not to go to action=edit?
[13:07:28] <Dantman>	Why?
[13:08:03] <Forrester>	hey... anyone around? :)=
[13:08:10] <Nikerabbit>	guess
[13:09:55] <AaronSchulz>	meh, GAID_FOR_UPDATE is easier
[13:12:20] <Forrester>	i got a question: Has a []-Link any influence on Special:UnusedImages?
[13:12:28] <Forrester>	do you got an answer?
[13:14:36] <CIA-50>	03aaron * r34611 10/trunk/extensions/FlaggedRevs/ (FlaggedArticle.php FlaggedRevs.php FlaggedRevsPage.php): 
[13:14:36] <CIA-50>	* Fix constant
[13:14:36] <CIA-50>	* Pass page_id to getRevisionTags()
[13:17:59] <MONK3000>	shaiaqua: I'm checking includes/Linker.php you could solve your problem there. I want it ns specific which may be harder
[13:18:22] <shaiaqua>	I fixed it, you could probab;y do the same with an if
[13:19:04] <shaiaqua>	ie if the namespace iis NS_USER or whatever, return ''
[13:19:11] <CIA-50>	03aaron * r34612 10/trunk/extensions/FlaggedRevs/FlaggedRevs.php: Uses a title now
[13:19:14] <MONK3000>	yep but i'm a noob
[13:19:22] <shaiaqua>	:p
[13:19:26] <MONK3000>	trial & error is hard
[13:20:58] <shaiaqua>	$wgArticle->getNamespace() is what you check against NS_USER, and you want to do this in linker.php line 242
[13:21:15] <shaiaqua>	if it is true, make $q = "";
[13:23:25] <MONK3000>	what does your line 242 read? i changed Linker.php before already ;)
[13:28:56] *Jack_Phoenix 	waves to ialex :-)
[13:29:14] <ialex>	hi Jack_Phoenix :)
[13:29:24] <Jack_Phoenix>	rar
[13:42:55] <wikibugs>	03(NEW) DPLforum extensions for Wikimedia Wikis - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14087 15enhancement; normal; Wikimedia: Site requests; (matthias.spam)
[13:45:41] <CIA-50>	03aaron * r34613 10/trunk/extensions/FlaggedRevs/FlaggedRevs.php: Slave lag
[13:59:45] <wikibugs>	03(mod) Install ConfirmAccount extension on enwiki - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13782  (10Thehelpfulonewiki)
[14:02:33] <MONK3000>	Where is Noarticletext called from? You know "There is currently no text in this page, you can" when the page doesn't exist. I think it's very buggy that a user page of an existing page looks the same as if the user name wouldn't exist. So I want MW to differ those cases and display different text if NS_USER
[14:03:01] <CIA-50>	03ialex * r34614 10/trunk/extensions/Newuserlog/Newuserlog.php: Kill useless extension function, simply define that variables when including the file
[14:03:15] <MONK3000>	**it's very buggy that an empty user page of an existing user looks the same as if the user name wouldn't exist
[14:04:47] <ialex>	MONK3000: you can install ParserFunctions and then use {{#ifeq:{{NAMESPACE}}|User|...
[14:06:03] <MONK3000>	on the MW message page?
[14:06:19] <ialex>	yes
[14:06:45] <ialex>	if you use 1.12, it works without any ugly hack
[14:06:46] <MONK3000>	uuh, solves 2 probs of me at once. didn't think of it
[14:06:57] <MONK3000>	yeah 1.12 :D
[14:12:14] <MONK3000>	works perfect ialex.
[14:13:48] <MONK3000>	although the same message appears now on User:NOTEXISTING and it still looks like the user exists
[14:15:04] <ialex>	maybe there's a {{#ifuserexists: function somewhere in an extensions
[14:15:44] <Simetrical>	We may as well have it, if we have #pageexists.
[14:15:47] <MONK3000>	well the problem I have is that most of my users have an empty user page
[14:16:32] <MONK3000>	and those user pages are treated like non existing users
[14:18:08] <Simetrical>	You can tell the difference, from things like contribution links in the corner.
[14:18:20] <Simetrical>	The connection between users and user pages is a little weird, it's true.
[14:18:27] <Simetrical>	Same with categories and category pages, images and image pages.
[14:18:47] <Simetrical>	They exist to describe some object or other, but their existence is independent of that object.
[14:19:31] <MONK3000>	yep that's not well solved. I mean for the User example: all users are listed in a nice table... ;)
[14:20:55] <MONK3000>	ok so how can I solve the Noarticletext message? atm i have it working like this {{#ifeq:{{NAMESPACE}}|User|existing user has no text|this page doesn't exist}}
[14:27:08] <MONK3000>	Wasn't there some possibilty to enter default text on a user page after registration? I would like to enter a message like "The user has not entered text yet".
[14:27:42] <Simetrical>	There's probably an extension somewhere.
[14:29:28] <MONK3000>	I tried hard searching it ;) "default user page" etc doesn't bring results. I asked that 3 times here already and nobody knew. But I know it exists as someone mentioned here half a year ago.
[14:33:20] <ialex>	for talk pages there is http://www.mediawiki.org/wiki/Extension:NewUserMessage
[14:34:54] <MONK3000>	ah ok i may be able to hack it to the user page ;)
[14:35:15] <ialex>	it should not be very difficult ;)
[14:37:12] <wikibugs>	03(mod) Support for global groups - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13773  (10Simetrical+wikibugs)
[14:39:45] <Zu>	hi, can anyone help me with a peculiar problem?
[14:40:28] <Zu>	I've just configured a new mediawiki installation but can't get file uploads going
[14:40:56] <Zu>	I repeatedly get "The file has no extension (like ".jpg")" when uploading a file
[14:41:55] <Zu>	The extension is correctly configured, I've toyed around with numerous settings (like allowing every file type, setting a different type detector, ...), /images/ has chmod 755, ...
[14:41:58] <Zu>	I don't know what to do next
[14:42:15] <CIA-50>	03aaron * r34615 10/trunk/extensions/FlaggedRevs/ (7 files): 
[14:42:15] <CIA-50>	* Revert back to r34539 (temporary)
[14:42:15] <CIA-50>	* Add tweak to avoid index miss
[14:43:57] <Zu>	test page: http://bankvandetoekomst.freehostia.com/wiki
[14:45:16] <shaiaqua>	my logo isn't showing up
[14:51:10] <MONK3000>	If getTalkPage() get's a User's talk page, what do I need to enter to get the normal User page?
[14:54:57] <CIA-50>	03aaron * r34616 10/trunk/phase3/includes/filerepo/File.php: Consistent type
[14:55:15] <wikibugs>	03(FIXED) FSRepo does not work well - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14084  +comment (10JSchulz_4587)
[15:01:58] <Dantman>	getUserPage() 
[15:02:45] <Dantman>	I suggest a good IDE and keeping a copy of MediaWiki handy
[15:03:27] <wikibugs>	03(ASSIGNED) Gender switch in user preferences - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13040  (10suidisz)
[15:04:36] <MONK3000>	yeah thanks. thought I tried that but did not .)
[15:27:21] <MONK3000>	is there a simple way to add a standard text on all existing but empty user pages? I'd prefer not using a bot since I'm too dumb for it
[15:28:06] <wikibugs>	03(mod) Gender switch in user preferences - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=13040  +comment (10azatoth)
[15:28:13] <Simetrical>	o_O
[15:28:28] <Simetrical>	Oh, i18n.
[15:28:29] <Simetrical>	Makes sense.
[15:28:35] <AzaTht>	Simetrical: to mine?
[15:28:57] <Nikerabbit>	Simetrical: except that if we add such magic word we effectively halv cache hit ratio :o
[15:29:12] <AaronSchulz>	AzaTht: I'd call it "ambigouos" ;)
[15:29:16] <AzaTht>	hehe
[15:29:25] <Simetrical>	AzaTht, yes.
[15:29:29] <Simetrical>	Well, to the bug in general.
[15:29:34] <Simetrical>	Nikerabbit, only if it's a magic word.
[15:29:35] <AzaTht>	ok
[15:29:39] <Simetrical>	Not if it only affects, say, default signature.
[15:29:54] <Simetrical>	"Unspecified" is what I use on my forum, that's the default.
[15:30:10] <AzaTht>	"unknown"
[15:30:36] <Nikerabbit>	Simetrical: they probably want that, to use it in interface translations
[15:30:55] <Simetrical>	Yes, that's true.
[15:31:09] <Simetrical>	Well, as long as it's not in the page text or sidebar or whatever, is there a problem?
[15:31:17] <Simetrical>	Interface text isn't cached if you're logged in, right?
[15:31:19] <Nikerabbit>	and then they want to use it in content too -> in any case doubles alternatives
[15:31:48] <Simetrical>	In content it's fine if it's substed.
[15:31:52] <Simetrical>	Just, not otherwise.  :)
[15:32:09] <Nikerabbit>	hmm difficult question
[15:33:16] <Nikerabbit>	perhaps it should be a per language setting which genders it has
[15:33:33] <Nikerabbit>	or uses
[15:33:40] <Nikerabbit>	most of them don't care
[15:33:40] <Simetrical>	Should you be allowed to specify "neuter" for languages that have that?  :)
[15:34:03] <Simetrical>	AFAIK, in most gendered languages there are conventions for addressing someone of unknown gender.
[15:34:13] <Nikerabbit>	Simetrical: don't know any languages thay have neuter "you"
[15:34:44] <AzaTht>	Nikerabbit: English?
[15:34:52] *Simetrical 	assigns one point to AzaTht 
[15:34:57] <Nikerabbit>	AzaTht: like?
[15:35:04] <AzaTht>	you
[15:35:07] <Simetrical>	The only languages that I know of that have neuter altogether are German, Greek, and Latin, I think.
[15:35:17] <Nikerabbit>	of course I meant neuter you in case there is male and female you
[15:35:18] <Simetrical>	That have gender altogether, that is.  :P
[15:35:28] <AzaTht>	hehe
[15:35:48] <Simetrical>	I think Latin has exclusively neuter second person, at least for pronouns.  "Tu" is for all genders, I'm pretty sure.
[15:35:51] <Nikerabbit>	that would be silly, because if tou speak to someone you ought to know his gender
[15:35:52] <Simetrical>	Buggered if I know for the others.
[15:36:07] <AzaTht>	Nikerabbit: non-modern swedish have neuter
[15:36:16] <AphelionZ>	hey everybody
[15:36:24] <Simetrical>	Hebrew is strictly male/female, English is ungendered, that's the end of my linguistic knowledge.
[15:37:14] <AzaTht>	Nikerabbit: http://en.wikipedia.org/wiki/Grammatical_gender#Masculine.2C_feminine.2C_and_neuter
[15:39:33] <Nikerabbit>	AzaTht: it is important to distinguish the grammatical gender in third person and in second person, because only the latter is what we talk now
[15:40:08] <AzaTht>	k
[15:43:16] <Nikerabbit>	hmm
[15:43:22] <Nikerabbit>	actually that's not true
[15:44:05] <Nikerabbit>	it is more like male/female words like king and queen
[15:44:32] <Nikerabbit>	complex things anyway
[15:45:50] <Nikerabbit>	anyway
[15:46:10] <Nikerabbit>	Simetrical: is it easy to domagic words that forcefully subst always?
[15:47:12] <Simetrical>	Nikerabbit, you would probably have to add a new special case to PST.
[15:47:24] <Simetrical>	Nikerabbit, but what you could do is just have them be cached.
[15:47:28] <Simetrical>	Even if that makes them meaningless.
[15:47:43] <Nikerabbit>	ugh
[15:47:49] <Simetrical>	That would make {{VIEWERGENDER}} fairly worthless, so you'd only want to use {{subst:VIEWERGENDER}} in practice.
[15:48:09] <Nikerabbit>	I'd rather return constantly something than random
[15:48:28] <Simetrical>	Well, it's possible to disable the parser cache, remember, in configuration.
[15:48:43] <Simetrical>	Of course it's not going to happen on Wikipedia, but for small third-party sites it might be totally tolerable.
[15:48:50] <Nikerabbit>	bah
[15:49:29] <wikibugs>	04(REOPENED) Kazakh message updates - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=12145  (10alefzet)
[15:50:18] <Nikerabbit>	maybe $wgAllowGrammarInContent which adds it to cache keys and enables it also when not parsing interface messages
[15:53:31] <wikibugs>	03(mod) Kazakh message updates - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=12145  +comment (10alefzet)
[16:06:37] <wikibugs>	03(mod) Default value for [[fa:MediaWiki:Sitesupport]] is English ( should be using MessagesFa.php) - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14061  (10huji.huji)
[16:08:09] <AphelionZ>	API Question - ive been querying index.php action = render and action = source and this seems to be really whats taking forever.. is there an alternative way to get this info?
[16:09:52] <Simetrical>	Bwahaha: http://www.abyssandapex.com/200710-wikihistory.html
[16:10:06] <Simetrical>	AphelionZ, what is it you're looking for?
[16:10:35] <AphelionZ>	ok so if i want to get the rendered version of an article I call index.php?action=render, and for source it's index.php?action=source
[16:10:49] <AphelionZ>	benchmarking shows that these calls take about 2 seconds each which is slowing my page load down to a crawl
[16:10:58] <Simetrical>	AphelionZ, that's because parsing pages is slow.
[16:11:06] <Simetrical>	You'll get the results faster if they're in the parser cache.
[16:11:12] <Simetrical>	But yeah, 2 seconds isn't surprising.
[16:11:18] <Simetrical>	Some enwiki pages take 15 seconds, or more.
[16:11:27] <Simetrical>	There have been a couple that take 60+ seconds, although those get fixed.
[16:11:30] <AphelionZ>	is the parser cache enabled automatically?
[16:11:33] <Simetrical>	Yes.
[16:11:36] <Simetrical>	But it's only a cache.
[16:11:45] <Simetrical>	It won't work if nobody else has viewed the same page, with the same preferences.
[16:12:13] <AphelionZ>	Is there anything I can do?
[16:12:32] <Simetrical>	Um, multithreading?  :)
[16:12:36] <AphelionZ>	heh
[16:12:52] <CIA-50>	03huji * r34617 10/trunk/phase3/languages/messages/MessagesFa.php: * Adding/updating Persian translations
[16:12:53] <Simetrical>	There's a reason Wikipedia's app servers are dual quad-core.
[16:12:57] <AphelionZ>	can I do that on a single server? Aaron Schulz told me to install APC
[16:13:10] <AphelionZ>	I'll try that next
[16:13:19] <Simetrical>	Your problem is as a site operator or as a bot operator?
[16:13:27] <Simetrical>	As a site operator, absolutely install APC, that's essential.
[16:13:33] <Simetrical>	Or similar (XCache, etc.)
[16:13:53] <AphelionZ>	site operator
[16:14:09] <Simetrical>	There might be a page somewhere on performance tuning.  Is it a large site?
[16:14:19] <nanotonne>	hello
[16:14:32] <Simetrical>	Something like APC is necessary for any PHP application to perform well.
[16:14:40] <AphelionZ>	its relatively large
[16:14:42] <AphelionZ>	http://dammit.lt/2007/01/26/mediawiki-performance-tuning/
[16:14:44] <Simetrical>	Specific MW config options I'm not sure of offhand, there are a lot of them.
[16:14:44] <nanotonne>	i need help with categorytree (in french if possible)
[16:14:53] <Simetrical>	You might consider things like $wgMiserMode.
[16:14:58] <AphelionZ>	MiserMode is on
[16:15:00] <nanotonne>	i have probleme with kavascript
[16:15:08] <Simetrical>	There are other things too, lots of config options scattered about.
[16:15:09] <nanotonne>	javascript
[16:15:40] <Simetrical>	AphelionZ, well, I'd trust anything written by Domas on MediaWiki performance.  :)
[16:16:04] <AphelionZ>	truth? or sarcasm?
[16:16:08] <Simetrical>	AphelionZ, truth.
[16:16:14] <AphelionZ>	excellent, thank you
[16:16:20] <AaronSchulz>	heh
[16:16:24] <Mike_lifeguard>	heresy! questioning domas' reliability :O
[16:16:29] <Simetrical>	He's the one who smacks everyone and disables stuff when we commit slow queries.
[16:16:31] <Simetrical>	:)
[16:16:46] *AaronSchulz 	is waiting for domas to return
[16:16:54] <AphelionZ>	me too, now
[16:17:27] <Mike_lifeguard>	AaronSchulz: do you have a moment to explain something to me regarding a setting in flaggedrevs?
[16:17:57] *Simetrical 	!asks Mike_lifeguard 
[16:18:01] <Mike_lifeguard>	I do not understand $wgFlaggedRevTags
[16:18:17] <Mike_lifeguard>	I am in the process of asking ;)
[16:18:54] <Mike_lifeguard>	the flag types that we are choosing the quality threshold for are ... what?
[16:24:14] <AaronSchulz>	you choose
[16:24:19] <AaronSchulz>	$wgFlaggedRevTags = array( 'accuracy'=>2, 'depth'=>1, 'style'=>1 );
[16:24:32] <AaronSchulz>	$wgFlaggedRevTags = array( 'accuracy'=>2, 'depth'=>1, 'wp1.0'=>1 );
[16:24:35] <AaronSchulz>	whatever
[16:24:49] <Mike_lifeguard>	so you can call them whatever you want
[16:24:55] <Mike_lifeguard>	and have as many or as few as you want
[16:24:57] <Mike_lifeguard>	?
[16:25:20] <Fire>	Hello.
[16:25:33] <MONK3000>	Is there some &action=save ?
[16:25:51] <AaronSchulz>	yeah
[16:25:59] *Mike_lifeguard 	loves that :DD
[16:26:09] <MONK3000>	so i call a URL and it saves e.g. a new article
[16:26:11] <AaronSchulz>	Simetrical: http://homepage.uvt.nl/~s485261/nbcmtp04.rm
[16:26:25] *AaronSchulz 	loves that
[16:26:30] <AaronSchulz>	always revs me up
[16:26:42] <Luxo>	MONK3000: no, you must send the data per post.
[16:27:05] <MONK3000>	:(
[16:27:23] <MONK3000>	ok got to edit 4000 empty user pages with some default text
[16:27:51] <MONK3000>	and I can't use a bot
[16:28:33] <Luxo>	MONK3000: why not?
[16:28:47] <Luxo>	(and why you must edit 4000 empty user pages..?)
[16:29:09] <MONK3000>	i'm too dumb to use a bot. command lines etc. are too much for me
[16:29:49] <Luxo>	MONK3000: then ask a bot-operator to do that. 
[16:29:53] <CIA-50>	03skizzerz * r34618 10/trunk/extensions/SimpleAntiSpam/ (. SimpleAntiSpam.i18n.php SimpleAntiSpam.php): Adding SimpleAntiSpam extension
[16:30:07] <MONK3000>	i must edit the user pages cos 90% of my users have blank user pages and I want to 1. get rid of the red links and 2. differ between non-existing users and empty pages of existing users
[16:30:42] <Mike_lifeguard>	MONK3000: There is an extension you may wish to look at.
[16:30:48] *Mike_lifeguard 	looks
[16:31:20] <MONK3000>	found Extension:Replace Text  but it's not working for big wikis
[16:31:29] <Mike_lifeguard>	http://www.mediawiki.org/wiki/Extension:NewUserMessage
[16:31:52] <MONK3000>	yes i use that NOW but it doesn't work for all the existign ones ;)
[16:33:02] <Mike_lifeguard>	true. well get a both on it then :)
[16:33:05] <Mike_lifeguard>	*a bot
[16:33:17] <MONK3000>	so what are bot-operators? 
[16:33:37] <Mike_lifeguard>	people who operate bots
[16:33:58] <MONK3000>	thought so. how can i get in touch with them?
[16:43:03] <Mike_lifeguard>	MONK3000: You may want to ask at #wikipedia-bag
[16:44:25] <MONK3000>	thanks
[17:00:02] <Luxo>	hm is there some CSS-class which is hidden in Internet Explorer but showed in the other browsers?
[17:07:12] <Skizzerz>	Luxo: not that I am aware of...
[17:07:46] <wikibugs>	03(mod) "You've got new messages" is displayed for a non-existent discussion page - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14077  +comment (10huji.huji)
[17:08:24] <Luxo>	Skizzerz: yeah I also don't found one.. thanks..
[17:09:39] <wikibugs>	03(mod) Power search form missing <label> for redirects check - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14063  +comment (10huji.huji)
[17:09:46] <MinuteElectron>	Notice:  Undefined variable: extensionsList in mediawiki/includes/SpecialUpload.php on line 1081
[17:11:14] <MinuteElectron>	Strict Standards:  Declaration of FileRepoStatus::newFatal() should be compatible with that of Status::newFatal() in mediawiki/includes/filerepo/FileRepoStatus.php on line 6
[17:11:18] <Nikerabbit>	MinuteElectron: you have $wgCheckFileExtensions = false?
[17:11:21] <MinuteElectron>	yes
[17:11:48] <Nikerabbit>	hrhm
[17:11:51] <Nikerabbit>	simple to fix
[17:12:15] <MinuteElectron>	probably, are you going to poke it -- or shall I?
[17:13:01] <Skizzerz>	hmm...
[17:13:43] *Skizzerz 	svn ups
[17:15:29] <CIA-50>	03minuteelectron * r34619 10/trunk/phase3/includes/SpecialUpload.php: Notice: Undefined variable: extensionsList in includes/SpecialUpload.php on line 1081
[17:15:49] <Nikerabbit>	Skizzerz: why are you inputting html directly? :o
[17:16:09] <Skizzerz>	Nikerabbit: for which part?
[17:17:24] <Nikerabbit>	Skizzerz: the extension
[17:17:26] <Skizzerz>	if it's for the Install function, that's displayed when the .php file is viewed directly and not through mediawiki. The other one is because I'm adding an HTML element to a form and as such would be stupid to parse some of it as wikitext (and thus cause unwanted line breaks) and some of it as HTML
[17:18:21] <Nikerabbit>	Skizzerz: well now it is even more stupid and inputting the message verbatim to it
[17:18:41] <Skizzerz>	if you want to fix it, go right ahead
[17:18:55] *Skizzerz 	doesn't see the issue though...
[17:19:49] <Nikerabbit>	unescaped output = bad
[17:19:53] <Skizzerz>	so they could put in a <script> tag? so what? sysops can edit Common.js, etc. anyway and add in a span to remote trigger javascript anyway
[17:20:14] <Nikerabbit>	Skizzerz: they can break xml-validity
[17:20:32] <Skizzerz>	{{sofixit}}
[17:21:09] <Skizzerz>	probably just requires adding a few letters after the wfMsg to make it parse as wikitext and not as unescaped html
[17:21:22] <CIA-50>	03minuteelectron * r34620 10/trunk/phase3/includes/filerepo/FileRepoStatus.php: Remove paramter that only appears to illustrate the fact more parameters can be added and causes an E_STRICT level error.
[17:21:54] <MinuteElectron>	yay!
[17:22:07] <Nikerabbit>	Skizzerz: well I don't care that much if you don't care
[17:22:22] <Nikerabbit>	just trying to tell the best practises
[17:22:41] <Nikerabbit>	but I don't have enough time to go and fix all extensions to follow them
[17:25:18] <Nikerabbit>	MinuteElectron: err that's bad, you must define it to avoid register globals
[17:25:49] <MinuteElectron>	Nikerabbit: =[
[17:26:02] <MinuteElectron>	DAMN YOU PHP, DAMN YOU
[17:26:18] <Nikerabbit>	MinuteElectron: just initialize it to '' or something
[17:26:46] <MinuteElectron>	Nikerabbit: Which one are we talking abouthere?
[17:26:58] <MinuteElectron>	the E_NOTICE ?
[17:27:05] <Nikerabbit>	MinuteElectron: r..619
[17:27:12] <MinuteElectron>	yeah
[17:27:22] <MinuteElectron>	What is extensionList actually existts?
[17:27:32] <Nikerabbit>	excuse me?
[17:27:43] <MinuteElectron>	sorry
[17:27:55] <MinuteElectron>	I mean, why is extensionList even not set in the first place?
[17:28:22] <Nikerabbit>	MinuteElectron: because there is ugly if ( X ) around the code that sets it
[17:28:29] <Nikerabbit>	and it is not initialised
[17:28:38] <Nikerabbit>	I wish php had const
[17:29:03] <CIA-50>	03skizzerz * r34621 10/trunk/extensions/SimpleAntiSpam/ (SimpleAntiSpam.i18n.php SimpleAntiSpam.php): wfMsg -> wfMsgExt so that raw HTML cannot be passed into form
[17:29:22] <MinuteElectron>	Nikerabbit: ok
[17:30:50] <Nikerabbit>	would be good case for Xml::inputLabel if it supported wikitext
[17:30:55] <Skizzerz>	Nikerabbit: better?
[17:31:05] <nazgjunk>	Hi, is there any way to create a list of users on a vanilla install rather than linking to the special page for that?
[17:31:30] <MinuteElectron>	Nikerabbit: Fixed.
[17:31:47] <nazgjunk>	create as in output it on a page
[17:31:50] <CIA-50>	03minuteelectron * r34622 10/trunk/phase3/includes/SpecialUpload.php: Avoid possible register_globals issue.
[17:33:39] <Nikerabbit>	Skizzerz: yeah, but doesn't label for=X need 'id' in the target?
[17:33:50] <Skizzerz>	no, it's based off of name=
[17:33:56] <MinuteElectron>	Is it?
[17:34:00] <MinuteElectron>	I'm sure label is based of id.
[17:34:01] <Skizzerz>	id is purely for linking/css purposes
[17:34:21] <MinuteElectron>	The fact that label was based off id rather than name is always something that has annoyed me.
[17:34:35] <Nikerabbit>	at least Xml:: says it refers to id
[17:34:54] <MinuteElectron>	Attribute Specifications  	
[17:34:55] <MinuteElectron>	* FOR=IDREF (associated form field)
[17:35:05] <Skizzerz>	it works anyway
[17:35:23] <Nikerabbit>	Skizzerz: not in firefox afaik
[17:35:32] <Nikerabbit>	clicking the text should activate the input
[17:35:36] <Skizzerz>	if I edit the HTML client-side to take away the display: none, the label displays properly with the input field
[17:35:50] <Skizzerz>	the whole purpose is NOT to fill in the input
[17:36:22] <Nikerabbit>	well, yeah
[17:36:35] <Skizzerz>	if it gets filled in, it aborts the save and gives a generic spam message
[17:39:21] *Skizzerz 	goes off to fix it since the validator doesn't like it though
[17:39:26] *Simetrical 	sighs at yet another totally incidental spam-stopping hack that only works as long as spambot authors don't spend half a second to look at it
[17:39:52] <Simetrical>	You know, you could probably hide it with CSS for all clients, if you aren't doing that.  Spambots don't use CSS, I'm sure.
[17:39:53] *G_SabinoMullane 	sighs in dismay at PHPs crappy attempt at date handling.
[17:40:07] *Simetrical 	likes to comment on things he hasn't actually looked at
[17:40:16] <Simetrical>	s/date handling/existing/
[17:41:27] <Skizzerz>	Simetrical: I just stole the basic HTML from Wikia's extension and reverse-engineered the PHP for it to make it open source. I do agree that it probably is quite ineffective, but it's there. Maybe if I changed the field to wpName or something to make it sound less like a trap...
[17:50:44] <Simetrical>	I don't think the bots will be using heuristics to analyze the form submission name to see if it's a trap.
[17:56:11] <Nikerabbit>	G_SabinoMullane: :o
[17:57:03] <Nikerabbit>	the date range is pretty limited
[17:57:55] <Simetrical>	The one awesome thing about PHP date handling is strtodate().
[17:58:03] <Simetrical>	Predictably, that functionality was not actually developed by the PHP team.
[17:58:32] <Simetrical>	Er, strtotime().
[17:59:08] <unstable>	I want to play flash video on my mediawiki page, what plugin do I want?
[17:59:30] <unstable>	There were different ones that could potentially do what I want. So I came here to get recommended a specific one.
[18:00:10] <Simetrical>	unstable, this isn't a great place for that.  It's mostly devs here, not users.  I'd just try one or two and see.
[18:00:50] <Nikerabbit>	Simetrical: ??
[18:01:12] <Nikerabbit>	I don't understand this flash-player thingie.. they're just ripping other sites most of the time
[18:10:05] <wikibugs>	03(mod) Transclusion should produce tree-like results in " What links here" - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=3241  (10rasamassen)
[18:14:40] <AerosAtar>	Hey guys, I'm having odd problems with the Tooltip extension, and I was hoping someone might be able to figure out why...
[18:15:45] <Nikerabbit>	maybe
[18:15:47] <AerosAtar>	Using the #tooltip parser through a template... the tooltip shoots off tot he left of the screen and cannot be moved, regardless of what I change the CSS to... any ideas why it might do that?
[18:16:17] <AerosAtar>	Weird thing is, on the Template: page, it appears where it should... it is only when I include the template that it goes weird...
[18:17:11] <AerosAtar>	I.e. Use the {{tooltip}} template on a {{weapon}} template.  Looking at the {{weapon}} template, it is fine.  When including {{weapon}} on a page, it shoots off to the left.
[18:17:41] <AerosAtar>	It _was_ working fine.  Then all of a sudden it went like this... without me changing anything...
[18:19:35] <AerosAtar>	Oddly, the Z-index wasn't working correcltly, either... but I've managed to fix that now...
[18:20:10] <Nikerabbit>	hmhm
[18:21:10] <AerosAtar>	(The navigation panel's Z-index:3 was being displayed above the Tooltip's Z-index:4... so I changed the navigation panel to 0... I couldn't see why the navigation panel was 3 anyway)
[18:22:33] <AerosAtar>	The z-index changes were made after it started going screy, so I know that wasn't the problem...
[18:22:44] <AerosAtar>	Any thoughts?
[18:22:56] <Nikerabbit>	nope
[18:24:31] <AerosAtar>	Well shoot...
[18:44:26] <CIA-50>	03skizzerz * r34623 10/trunk/extensions/SimpleAntiSpam/SimpleAntiSpam.php: adding id to input for label to function properly
[18:54:29] <unstable>	Nikerabbit: We run a LUG website, and record our talks. So we wanted to make it easy for people to watch the talks from our website.
[18:56:24] <unstable>	Nikerabbit / Simetrical what flash plugin to use? do you have a url?
[18:57:33] <AphelionZ>	any SQL ninjas around?
[18:58:12] <AaronSchulz>	hm?
[18:58:26] <AaronSchulz>	"ninjas" puts up a pretty high standard ;)
[18:59:05] *Fire 	wonders if AaronSchulz rather has auto-highlight for "SQL" or for "ninja".
[19:01:00] <AphelionZ>	haha Aaron, I dont want to bug you but I've been stupted on this for three days now
[19:01:13] <AphelionZ>	've got this nasty query http://pastebin.com/d48f8bd4e and the problem is in the join with the postmeta table. I'm trying to match the keyword for an advanced search and joining these tables should produce more results but it still produces the same
[19:01:16] <AphelionZ>	I*ve
[19:01:51] *AaronSchulz 	whistles in aww
[19:02:14] <AaronSchulz>	AphelionZ: just for kicks, show that query to domas
[19:02:22] <wikibugs>	14(DUP)  Image links that points to redirects on Commons returns a blank page on local wiki - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14078  +comment (10Bryan.TongMinh)
[19:02:28] <AaronSchulz>	anyway, is it wrong, empty, or errors out?
[19:02:30] <wikibugs>	03(mod) Image redirect implementation is very confusing - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13943  +comment (10Bryan.TongMinh)
[19:02:35] <AphelionZ>	nope it works
[19:02:50] <AaronSchulz>	so what is the problem again?
[19:02:58] <AphelionZ>	but it should return 11 or 12 results (the postmeta results) instead of the 6 that it returns
[19:03:21] <AphelionZ>	its anchored by the wrong table maybe? i tried LEFT RIGHT INNER OUTER JOINS
[19:03:25] <AaronSchulz>	so then it is wrong ;)
[19:03:28] <AphelionZ>	even the hidden 'SPACE JOIN'
[19:03:45] <AphelionZ>	ok then its wrong haha
[19:03:52] <AaronSchulz>	hmm, hard to tell what the query is trying to do
[19:03:56] *AaronSchulz 	has to think
[19:05:12] <AphelionZ>	ok the joins are just letting it search the keyword against a few different tables.. all of the other tables return single rows except for the postmeta table, which is the reason for the "WHERE" bit on that join
[19:05:37] <AphelionZ>	but if i just search the postmeta table for the keyword, 1112 results come up, but this query is only returning what it was returning before that join was in, which is 6
[19:06:55] <wikibugs>	03(mod) Power search form missing <label> for redirects check - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14063  (10siebrand)
[19:07:56] <AaronSchulz>	well, I'd move "INNER JOIN wp_postmeta tr3 ON wp_posts.ID = tr3.post_id" after "FROM wp_posts"
[19:08:05] <AaronSchulz>	that makes it clearer
[19:08:12] <AphelionZ>	k
[19:08:31] <AaronSchulz>	so the INNER JOIN wp_postmeta is the problem, right?
[19:08:48] <AphelionZ>	yeah, i'd say so, could be the corresponding MATCH AGAINST
[19:08:52] <AphelionZ>	but its probably the JOIN
[19:09:08] <AphelionZ>	i thought about making the query based on postmeta first and then joining the posts table
[19:09:57] <AaronSchulz>	without that JOIN, are the results right?
[19:11:38] <AphelionZ>	yeah
[19:11:50] <AphelionZ>	and they're still fine, just missing a few
[19:12:26] <AaronSchulz>	do you only want rows that have the wp_postmeta fields?
[19:13:10] <AphelionZ>	i want the rows that match in every table, regardless of whether or not they match in the others
[19:13:18] <AphelionZ>	in each* table, i should say
[19:15:15] <AaronSchulz>	then that should be a LEFT JOIN
[19:16:19] <AphelionZ>	thats what I thought but it still doesnt work
[19:16:27] <AphelionZ>	in fact, that's what I tried first!
[19:17:02] <AaronSchulz>	go back to doing that
[19:17:05] <AaronSchulz>	there must be another bug
[19:17:10] <AaronSchulz>	what is wrong when you do that?
[19:18:58] <AphelionZ>	same results
[19:20:23] <AphelionZ>	6 instead of 11
[19:28:29] <AaronSchulz>	why does it say "wp_postmeta tr3"
[19:28:39] <AaronSchulz>	ohh, nvm
[19:30:57] <AaronSchulz>	AphelionZ: are any wp_postmeta fields in the WHERE?
[19:32:55] <AphelionZ>	hey im back, sorry
[19:32:57] <AphelionZ>	got disconnected
[19:33:03] <AphelionZ>	AaronSchulz: any ideas?
[19:33:23] <AaronSchulz>	well the join should be LEFT, the issue is what else is wrong
[19:33:34] <AaronSchulz>	[15:30]	<AaronSchulz>	AphelionZ: are any wp_postmeta fields in the WHERE?
[19:33:48] <AphelionZ>	they're in the MATCH AGAINST below
[19:34:28] <AphelionZ>	the tr3.meta_value
[19:34:34] <AaronSchulz>	I'd make the ON clause with the AND in () as well
[19:34:39] <AaronSchulz>	to be cleaner
[19:34:50] <AphelionZ>	k let me do your changes and try it and make a new paste
[19:37:47] <AphelionZ>	http://pastebin.com/d51fd5e85
[19:39:56] <AaronSchulz>	what are the wp_postmeta fields ?
[19:40:21] <AphelionZ>	meta_id, post_id, meta_key, meta_value
[19:40:28] <AphelionZ>	and each post is associated with any number of meta_keys
[19:40:38] <AphelionZ>	and i tried to narrow it to just the 'destination' keys
[19:41:10] <AaronSchulz>	what is the query when you don't join on wp_postmeta ?
[19:41:53] <AphelionZ>	http://pastebin.com/d589f430a
[19:42:28] <AaronSchulz>	and why do you need DISTINCT?
[19:42:44] <Nikerabbit>	hello
[19:43:03] <AphelionZ>	maybe I dont let me try without it
[19:43:16] <AaronSchulz>	do not just add DISTINCT "to be sure"
[19:43:21] <AaronSchulz>	either you need it or you don't
[19:43:27] <AaronSchulz>	it is really not very fast
[19:43:54] <AaronSchulz>	http://pastebin.com/d589f430a
[19:43:59] <AaronSchulz>	AS score, tr3.meta_value ?
[19:44:02] <AaronSchulz>	where is tr3 set?
[19:44:06] <AphelionZ>	oops
[19:44:18] <AphelionZ>	just forgot to take that out when you asked me to remove the postmeta stuff
[19:44:31] <AaronSchulz>	in fact, mixing DISTINCT and GROUP BY is really funky
[19:44:35] <AphelionZ>	without distinct is WAY faster but doesnt solve the problem
[19:44:56] <AaronSchulz>	well, it solves slowness ;)
[19:45:01] <AphelionZ>	totally!
[19:46:30] <AphelionZ>	ugh this is killing me
[19:47:18] <AaronSchulz>	run it with and without the JOIN
[19:47:26] <AaronSchulz>	with the distinct stuff gone
[19:49:21] <CIA-50>	03btongminh * r34624 10/trunk/phase3/includes/ (Article.php ImagePage.php): Put code to view redirect page in new method Article::viewRedirect in order to allow displaying redirects on foreign file repos.
[19:50:01] <AphelionZ>	its all the same, pretty much
[19:50:07] <AphelionZ>	except its faster with the distinct
[19:50:16] <AphelionZ>	i tried it with postmeta in the initial join
[19:50:24] <AphelionZ>	i mean, in the initial FROM
[19:50:28] <AphelionZ>	and postmeta in the JOIN
[19:50:38] <AphelionZ>	er.. and posts* in the join
[19:50:44] <AaronSchulz>	ok
[19:51:03] <AphelionZ>	same
[19:51:08] <AphelionZ>	so maybe the problem is in the WHERE
[19:51:20] <AphelionZ>	i could try replacing the MATCH AGAINST with a simple LIKE statement maybe
[19:52:39] <AphelionZ>	http://pastebin.com/d3c7f4461
[19:52:42] <AphelionZ>	same results :-/
[19:53:39] <AaronSchulz>	wp_posts.post_title, wp_posts.post_excerpt, wp_posts.post_name are not aggregated either
[19:54:03] <AphelionZ>	meaning they should be t1 or t0 or something?
[19:54:22] <AaronSchulz>	as well as score, tr3.meta_value
[19:54:35] <AaronSchulz>	AphelionZ: they need to make sense with the GROUP BY
[19:54:45] *AaronSchulz 	is surprised that works at all :)
[19:55:08] <Fire-away>	have a nice evening/morning/whatever
[19:55:33] <AaronSchulz>	heh
[19:55:45] <AphelionZ>	whats wrong with the group by?
[19:56:15] <AaronSchulz>	AphelionZ: do any of the tr1,tr2, and tr3 tables have multiple rows that correspond to the same post ID ?
[19:56:27] <AphelionZ>	yes, tr3 does
[19:56:43] <AphelionZ>	well, it shouldnt because of the AND meta_key = 'destination'
[19:56:57] <AphelionZ>	but each post has several pieces of metadata attached to it, yes
[19:56:58] <AaronSchulz>	then the GROUP BY is not needed
[19:57:16] <AaronSchulz>	if post ID, meta_key is a key of tr3
[19:57:22] <AaronSchulz>	* (post ID, meta_key)
[19:57:36] <AaronSchulz>	and by "key", I mean "unique key" ;)
[19:59:43] <AphelionZ>	the primary key is meta_id
[20:00:00] <AphelionZ>	and then it has post_id as an index and meta_key as an index
[20:00:46] <AphelionZ>	haha taking out the group by gives me six identical results
[20:00:47] <AaronSchulz>	well, as long as it effectively is unique, then the GROUP BY can go
[20:01:03] <AaronSchulz>	buy you should add a unique key
[20:01:17] <AaronSchulz>	AphelionZ: can you do SHOW TABLE for tr3?
[20:01:23] <AphelionZ>	sure thing
[20:02:59] <AphelionZ>	umm do you mean SHOW COLUMNS ? show table is throwing an error
[20:03:12] <AaronSchulz>	that's fine
[20:03:28] <AaronSchulz>	but (post_id, meta_key) *should* be unique right?
[20:04:22] <AphelionZ>	http://pastebin.com/d1fc680b7
[20:04:26] <AphelionZ>	yeah it should be
[20:04:44] <AaronSchulz>	remove the post_id index
[20:04:53] <AaronSchulz>	replace it with UNIQUE KEY (post_id,meta_key)
[20:06:28] <AaronSchulz>	AphelionZ: actually, what are all the current indexes?
[20:06:45] <AaronSchulz>	"post_id	bigint(20)	NO	MUL	0	" :s
[20:07:46] <AphelionZ>	http://pastebin.com/d45eec9f0
[20:08:27] <AaronSchulz>	do SHOW INDEX FROM ?
[20:09:48] <AphelionZ>	http://pastebin.com/d29dfd33d
[20:10:13] <AaronSchulz>	looks like you have PRIMARY on meta_key, and then an extra meta_key index
[20:10:16] <AaronSchulz>	I'd kill that
[20:10:26] <AaronSchulz>	and replace the post_id one with UNIQUE KEY (post_id,meta_key)
[20:10:41] <AphelionZ>	no meta_id and meta_key are two different things
[20:12:00] <AaronSchulz>	oh, I see, looked the same to me
[20:12:11] <AaronSchulz>	ok, did you redo post_id yet?
[20:12:22] <AphelionZ>	not yet hang on
[20:13:27] <AphelionZ>	#1062 - Duplicate entry '45-_wp_page_template' for key 2 
[20:13:48] <AaronSchulz>	so either there is a wrong entry or the thing is not intended to be unique
[20:13:57] <AphelionZ>	ughhhh
[20:13:59] <AaronSchulz>	not really a mysql issue
[20:14:24] <AaronSchulz>	AphelionZ: if you meant for it to be unique, then you messed up somewhere :)
[20:14:35] <AphelionZ>	fair enough
[20:14:40] <AphelionZ>	i just want this stupid thing to work
[20:14:47] <AphelionZ>	how do I put the original index back
[20:15:24] <AaronSchulz>	AphelionZ: do SELECT DISTICT post_id,meta_key
[20:15:32] <AaronSchulz>	I wan't to see how many dups there are
[20:16:04] <AaronSchulz>	anyways, to add the old index, do ALTER TABLE ADD INDEX post_id (post_id)
[20:16:53] <AaronSchulz>	ALTER TABLE `wp_postmeta`ADD INDEX post_id (post_id)
[20:17:22] <AphelionZ>	that select distinct returns 2,724 rows
[20:17:24] <Charitwo>	aha
[20:17:58] <AaronSchulz>	AphelionZ: ok, now do SELECT COUNT(*) WHERE meta_key = 'destination'
[20:18:40] <AphelionZ>	385
[20:18:44] <AphelionZ>	which makes sense
[20:19:57] <AaronSchulz>	AphelionZ: SELECT DISTINCT post_id,meta_key WHERE meta_key = 'destination'
[20:20:26] <AaronSchulz>	heh, actually that very first distinct was not helpful ;)
[20:20:37] <AphelionZ>	383
[20:20:41] <AaronSchulz>	so 2 dups
[20:20:58] <AphelionZ>	yeah they had their interns putting this stuff in
[20:21:06] <AphelionZ>	so i could easily see them putting a dupe in
[20:21:11] <AaronSchulz>	I'd kill the dups, then add the unique key
[20:21:26] <AphelionZ>	the dupes are on other things besides destinations though
[20:21:34] <AphelionZ>	and besides will that really help my original query?
[20:21:43] <AaronSchulz>	it would kill the group by
[20:21:49] <siebrand>	PHP Notice:  Use of undefined constant FOR_UPDATE - assumed 'FOR_UPDATE' in /var/www/w/includes/Title.php on line 200
[20:22:38] <CIA-50>	03greg * r34625 10/trunk/phase3/includes/SpecialContributions.php: (log message trimmed)
[20:22:38] <CIA-50>	Clean up date formatting per bug 14067, remove assumptions that things like
[20:22:38] <CIA-50>	20051300000000 will work. It's fine for dates-as-integers, but goes horribly
[20:22:38] <CIA-50>	wrong when fed to $dbr->timestamp(), which calls gmmktime(). The latter function
[20:22:38] <CIA-50>	has a bug/feature causing any 0s in the month or day field to cause a month
[20:22:41] <CIA-50>	or day's *subtraction*. We might want to modify timestamp() to handle that, as
[20:22:43] <CIA-50>	I can imagine any MW code would want to rely on such a feature, but that's not
[20:22:51] <AphelionZ>	hmm
[20:22:53] <siebrand>	PHP Notice:  Use of undefined constant FOR_UPDATE - assumed 'FOR_UPDATE' in /var/www/w/includes/Title.php on line 200 (in case it was missed :))
[20:23:02] <CIA-50>	03aaron * r34626 10/trunk/phase3/includes/Title.php: GAID_FOR_UPDATE
[20:24:10] <wikibugs>	03(FIXED) Invalid timestamp in SpecialContributions.php - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14067  +comment (10greg)
[20:24:27] <CIA-50>	03aaron * r34627 10/trunk/phase3/includes/Title.php: Tweak doc
[20:24:51] <AphelionZ>	the problem from the user point of view is that if they type in a destination name, like Paris, not all the results are showing up because its not searching within that table
[20:25:00] <AaronSchulz>	AphelionZ: hmm, let me think of a nice dup finder
[20:26:26] <AphelionZ>	k
[20:27:26] <AaronSchulz>	SELECT post_id,meta_key,COUNT(*) as count FROM `wp_postmeta` GROUP BY post_id,meta_key HAVING count > 1
[20:27:47] <AphelionZ>	103 dups
[20:28:00] <AaronSchulz>	can you see the list? they are dups right?
[20:28:07] <AphelionZ>	i can't delete these though
[20:28:17] <AphelionZ>	unfortunately, they're not supposed to be unique
[20:28:37] <AaronSchulz>	bah :)
[20:29:04] <AphelionZ>	which brings us back to: http://pastebin.com/d3c7f4461
[20:29:08] <AaronSchulz>	so you could have two "destination"s
[20:29:25] <AphelionZ>	you *could* but that would be a mistake... and we can assume there's just one as far as 'destination' goes
[20:29:41] <AaronSchulz>	AphelionZ: what other meta things are there?
[20:29:50] <AaronSchulz>	aside from "destination"
[20:29:52] <AphelionZ>	keyword, photo url, photo credit
[20:29:59] <AaronSchulz>	should be unique
[20:30:10] <AaronSchulz>	otherwise, it's poor db layout
[20:30:24] <AphelionZ>	true, true
[20:30:26] <AaronSchulz>	which makes queries like the one you are trying to get harder and mesier
[20:30:35] <AaronSchulz>	it comes back to bite you in the rera
[20:30:37] <AaronSchulz>	*rear
[20:30:39] <AphelionZ>	haha
[20:30:47] <AphelionZ>	when all i want is simply *more* results
[20:39:55] <CIA-50>	03greg * r34628 10/trunk/phase3/includes/GlobalFunctions.php: 
[20:39:55] <CIA-50>	Refactor duplicated code into one area.
[20:39:55] <CIA-50>	Remove unneeded parens in regexes, don't set matches array if not needed.
[20:39:55] <CIA-50>	Add warning about odd gmmktime() behavior.
[20:48:59] <Dragonfly6-7>	is there a way to go through the deletion logs and find all deleted articles whose names begin with a given string?
[21:01:12] <AerosAtar>	Anyone have any idea why putting a tooltip tag/parser (from the Tooltip extension) inside a wiki table would screw with it?  It seems to be drawing the style elements from the table rather than its own CSS...
[21:02:25] <AerosAtar>	Applying the styles to the contents, as well... even when the content is drawn from another page with different styles.
[21:02:43] <CIA-50>	03siebrand * r34629 10/trunk/phase3/languages/messages/ (31 files): Localisation updates for core messages from Betawiki (2008-05-11 22:57 CEST)
[21:05:43] <CIA-50>	03siebrand * r34630 10/trunk/extensions/ (49 files in 49 dirs): Localisation updates for extension messages from Betawiki (2008-05-11 22:57 CEST)
[21:07:40] <wikibugs>	03(FIXED) Messages from i18n file does not appear in Special:Allmessages - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=12279  +comment (10raimond.spekking)
[21:07:46] <CIA-50>	03btongminh * r34631 10/trunk/phase3/includes/SpecialEmailuser.php: Refactor SpecialEmailUser.php to split outputting from other stuff so that a to be written API module can easily use the existing code.
[21:09:14] <MONK3000>	has anyone found a solution for the UNIQ ... QINU problem? I would like to install Extension:CurrentUsers but can't since it destroys all following inserts on a page
[21:23:44] <AaronSchulz>	who needs pages?
[21:31:17] <CIA-50>	03aaron * r34632 10/trunk/extensions/FlaggedRevs/ (7 files): Scap done already - bring back commits reverted in r34615
[21:36:24] <wikibugs>	03(mod) Disable images in signatures on enwiki - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=6379  (10compwhizii)
[21:36:51] <CrazyTB>	How can I add system-wide styles (css) to my mediawiki?
[21:37:07] <Skizzerz>	!css
[21:37:07] --mwbot--	To change styles for your wiki, go to one of the MediaWiki:xxx.css wiki page and put your custom styles there (sysop/admin rights required). MediaWiki:common.css is for all skins and should be used for content styles. MediaWiki:monobook.css is for the MonoBook skin (default), etc. For more information, see !skins and <http://www.mediawiki.org/wiki/Manual:Skin_configuration>
[21:37:50] <CrazyTB>	yhanks
[21:37:59] <CrazyTB>	I thought I needed to hand-edit the css files at server
[21:40:22] <CrazyTB>	Is there any wiki format for marking up a quotation?
[21:43:29] <CIA-50>	03nikerabbit * r34633 10/trunk/phase3/includes/ (SpecialNewpages.php SpecialWhatlinkshere.php Xml.php): * Code cleanup
[21:43:50] <Splarka>	what type of quotation? quotation marks, <blockquote>, <pre>, <poem>, what?
[21:44:10] <CrazyTB>	blockquote
[21:48:10] <Splarka>	well, wikicode works in a blockquote fine
[21:48:28] <Splarka>	<blockquote>''foo'' '''bar''' baz <span style="color:red;">foo</span> [[bar]] <sup>baz</sup></blockquote>
[21:48:36] <Splarka>	as does whitelisted html
[21:48:40] <CrazyTB>	ok, thanks
[21:49:39] <Splarka>	wikicode won't work in certain tags, mostly tags that are parser hook extensions like <poem> or <pre>, rather than html or transclusion magic... like <span> or <includeonly>
[21:55:57] <CIA-50>	03nikerabbit * r34634 10/trunk/phase3/includes/SpecialNewpages.php: * Forget the offset in the form also
[22:52:52] <wikibugs>	03(mod) Invalid timestamp in SpecialContributions.php - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14067  (10N/A)
[23:08:30] <agentdcooper>	hoping someone can assist me, I am working on installing/configuring mediawiki+lucene for full text search capability
[23:08:59] <agentdcooper>	i got lucene installed, and Extension:MWSearch so far i think
[23:09:37] <agentdcooper>	for some reason my wikidb.xml or the index's are not being created properly
[23:10:05] <agentdcooper>	this has to do with how the "lsearch daemon" interects with mediawiki
[23:10:44] <agentdcooper>	wondering if anyone has done this before, re: mediwiki+lucene (for full text search)?
[23:11:13] <flyingparchment>	i guess i've done it before, since i wrote the original lucene search, but i'm pretty sure it's changed entirely since then ;)
[23:11:35] <agentdcooper>	i am getting the error: 259252 [pool-2-thread-3] ERROR org.wikimedia.lsearch.search.SearchEngine  - Internal error in SearchEngine trying to make WikiSearcher: /usr/local/search/indexes/search/wikidb/segments (No such file or directory)
[23:11:48] <HeyJued>	Hello I have a quick question. How do I get my Text to format correctly and not run under pictures or infoboxes?  http://leetpwned.com/index.php?title=BioShock  I would like the text to format like http://en.wikipedia.org/wiki/BioShock
[23:12:54] <Smaug>	HeyJued: you want your text to not be in an infobox at all?
[23:13:43] <agentdcooper>	when "starting up" lsearchd it looks like it is gonna work, but comes up to:: 2724 [main] ERROR org.wikimedia.lsearch.search.SearcherCache  - I/O Error opening index at path /usr/local/search/indexes/search/wikidb : /usr/local/search/indexes/search/wikidb/segments (No such file or directory)  and 2747 [Thread-2] INFO  org.wikimedia.lsearch.frontend.HTTPIndexServer  - Started server at port 8321
[23:13:43] <agentdcooper>	2749 [main] ERROR org.wikimedia.lsearch.search.SearcherCache  - I/O Error opening index at path /usr/local/search/indexes/search/wikidb : /usr/local/search/indexes/search/wikidb/segments (No such file or directory)
[23:13:57] <HeyJued>	I dont want the paragraph of the actual article to run under the infobox. While keeping the text in the info box.
[23:14:06] <HeyJued>	Similar to a wordwrap
[23:14:10] <Charitwo>	agentdcooper: you might want to check out   http://rafb.net/paste/ for showing large amounts of text, rather than pasting it to the chanel
[23:14:26] <agentdcooper>	Charitwo: perfect
[23:15:18] <HeyJued>	Smaug: You can see the difference between my wiki and Wikipeidas http://en.wikipedia.org/wiki/BioShock    
[23:15:29] <HeyJued>	My Wiki ---> http://leetpwned.com/index.php?title=BioShock
[23:16:35] <agentdcooper>	ok, so the error i am getting is @ http://rafb.net/p/aVNXav52.html  --- that is what happens when I start lsearchd, which is Extension:Lucene-search tutorial talks about running/installing in the very first step
[23:17:00] <agentdcooper>	i just cannot get LSearch daemon to create the 'indexes' =(
[23:17:06] <agentdcooper>	for my mediawiki site
[23:18:34] <Smaug>	HeyJued: I've edited the page.  Is this what you want?
[23:19:16] <agentdcooper>	so, when I tell it to build my index -- which I'd think it's looking for when lsearch daemon starts up -- the line "php maintenance/dumpBackup.php --current --quiet > wikidb.xml &&   java -cp LuceneSearch.jar org.wikimedia.lsearch.importer.Importer -s wikidb.xml wikidb" comes back with no errors
[23:19:17] <HeyJued>	Smaug: I was trying to get a similar format to this page http://en.wikipedia.org/wiki/BioShock
[23:20:14] <agentdcooper>	my thought is the "lsearch" deamon is not accessing my true mediwikiDB properly, hence it cannot find anything to index?
[23:20:52] <Smaug>	HeyJued: you could take a look at http://en.wikipedia.org/wiki/Help:Tables and try to duplicate it
[23:21:16] <Smaug>	i don't know the syntax for putting a table on the right in my head
[23:22:48] <HeyJued>	Thanks you for the link. Do you know what it is called? (wordwrapg)
[23:23:24] <Smaug>	what what is called?
[23:24:56] <HeyJued>	I guess its text formatting? 
[23:25:56] <Smaug>	table formatting, text formatting
[23:26:52] <HeyJued>	Thank you
[23:28:19] <Smaug>	yw :)
[23:34:25] <agentdcooper>	haha
[23:34:28] <agentdcooper>	found my fix
[23:34:34] <agentdcooper>	@ http://www.mediawiki.org/wiki/Extension_talk:LuceneSearch
[23:34:37] <agentdcooper>	w00t
[23:34:38] <agentdcooper>	peace