[00:09:49] <CIA-6>	03rainman * r31134 10/branches/lucene-search-2.1/ (28 files in 10 dirs): Fuzzy queris and suggestions over all namespaces
[00:11:10] <Giszmo>	hi
[00:11:10] <Giszmo>	how do i google for mediawiki-features without hitting tons of wikis?
[00:11:10] <Giszmo>	right now i'm searching for some way to count occurances of a symbol in a page. for example like this:
[00:11:10] <Giszmo>	;voted pro/contra: {{count|pro}}{{count|contra}}
[00:11:10] <Giszmo>	{{pro}} me, {{pro}} her, {{contra}} him
[00:19:44] <BrokenArrow>	Giszmo: other than limiting your searches to domains like mediawiki.org and wikimedia.org (many things arestill on meta.wikimedia.org for historical reasons) I don't see whatto do
[00:38:39] <Giszmo>	thanx BrokenArrow. didn't find much though
[00:38:46] <Giszmo>	cu
[01:22:10] <wikibugs>	03(NEW) trackback.php XML parse error? - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13086 15enhancement; normal; MediaWiki: General/Unknown; (dsimonto)
[01:59:05] <Mitchazenia>	I'm trying to set up a wiki, yet something in the setup form (with the passwords) are messing up
[01:59:13] <Mitchazenia>	can someone help?
[02:04:02] <linuxpoet>	btw guys, thanks for linking my article in mediawiki that was a surprise
[02:04:38] <Mitchazenia>	anyone?
[02:31:37] <wikibugs>	03(mod) Special page listing changes to all pages linking *to* a given page - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=6528  +comment (10spidermannequin)
[02:39:55] <wikibugs>	03(NEW) Organize Whatlinkshere by section ( and support transclusion) - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13087 15enhancement; normal; MediaWiki: Templates; (spidermannequin)
[02:48:47] <Wiredtape>	does anyone know of a text parser that searches for wildcard expression and outputs case to xml? for instance <div id="case">*</div> | * could be additional tags or other things and so on.. 
[02:52:32] <Skizzerz>	Wiredtape: is this for PHP or JavaScript?
[02:53:16] <Wiredtape>	i have a bunch of txt/html files that I want to scan through and output information based on what i find in the case
[02:53:38] <Skizzerz>	ok
[02:53:53] <Wiredtape>	any chance you know of anything like that? :)
[02:53:56] <Skizzerz>	use regex ;)
[02:55:28] <Wiredtape>	:) 
[02:56:10] <Skizzerz>	preg_match() for PHP, node.match() for javascript, no idea what for other languages...
[02:57:15] <Wiredtape>	yeah.. only I would need to create quite a wildcard facility for that.. as preg_match(blablabla*blablab) won't work afaik..
[02:57:47] <Skizzerz>	.*
[02:58:04] <Skizzerz>	preg_match(blablabla.*blablab)
[02:58:11] <Wiredtape>	would that work?
[02:58:15] <Skizzerz>	yes
[02:58:26] <Wiredtape>	hmm.. let's see, thanks Skizzers :)
[02:58:31] <Skizzerz>	. matches any one character, * says zero or more of the previous thing
[02:58:57] <Skizzerz>	.* therefore matches any number of any combination of characters, leaving only enough to finish matching everything after
[02:59:39] <Wiredtape>	is there any way for me to get .* afterwards?
[02:59:47] <Wiredtape>	or whatever was in there?
[02:59:58] <Skizzerz>	yeah, surround it in ()
[03:00:05] <Wiredtape>	and then $1?
[03:00:09] <Skizzerz>	although you'd probably want it lazy then
[03:00:30] <Skizzerz>	Wiredtape: not in PHP
[03:00:33] <Wiredtape>	lazy?
[03:02:22] <Skizzerz>	yes, preg_match('aaa(.*)b', 'aaaaaaaaabbbbababaaaab', $matches) would give 'aaaaaabbbbababaaaa' as your first group, but preg_match('aaa(.*?)b', 'aaaaaaaaabbbbababaaaab', $matches) would only give out 'aaaaaa'
[03:02:50] <Skizzerz>	http://us.php.net/preg_match for more info on calling the captured bits
[03:03:01] <Wiredtape>	you mean $matches[0] would have 'aaaaaa'?
[03:03:14] <Skizzerz>	no, $matches[1] would
[03:03:15] <Wiredtape>	I'm actually looking at it right now.. 
[03:03:34] <Skizzerz>	$matches[0] would contain the entire string
[03:03:52] <Wiredtape>	ok.. now, does preg_match look at spaces as well?
[03:04:41] <Skizzerz>	yes
[03:06:01] <Wiredtape>	Skizzerz, thanks! one last q, where can i find the different expressions preg_match uses? (like .*)..
[03:06:38] <Skizzerz>	http://en.wikipedia.org/wiki/Regular_expression seems like a good source
[03:06:53] <Wiredtape>	great.. thanks a lot :)
[03:07:59] <Skizzerz>	(you're|ur) (very)? ?welcome (:\) )+
[03:08:07] <Wiredtape>	:) lol
[03:16:33] <xtine>	Wiredtape: try using str_replace, strpos, or substr
[03:16:41] <xtine>	Wiredtape: preg_match can get really slow if you call it a lot
[03:17:23] <Wiredtape>	xtine, I looked into those earlier.. can they handle expressions as well?
[03:17:50] <xtine>	well you want to run a find and replace right?
[03:18:08] <Wiredtape>	with wildcards.. and to be able to access those wildcards..
[03:18:18] <xtine>	you can mix it up so you can set what you want to do
[03:18:32] <xtine>	it's just if you are worried about speed
[03:18:46] <xtine>	if this is just for a small thing, no worries
[03:19:08] <xtine>	but if you are doing a lot of text crunching, preg_replace can become very expensive
[03:19:38] <Wiredtape>	so you're saying make string searches and the preg_match as a case option? this is a good idea, though i'm not really worried about speed , this should run on my local machine and not on the server.
[03:20:13] <Wiredtape>	btw, anyone know where I can find info about the xml model i need to make for special:import to work?
[03:21:27] <Wiredtape>	xtine, thanks :)
[03:21:35] <xtine>	nps
[03:21:56] <xtine>	i had a web dev friend harp on me how it's not good practice to just call preg_replace when you can use other more efficient functions to essentially do the same thing
[03:22:16] <Wiredtape>	:) 
[03:22:23] <xtine>	of course i don't know what level you are working at
[03:22:47] <Wiredtape>	i think this is the "i dont care" level :)
[03:22:50] <xtine>	lol
[03:23:20] <Wiredtape>	i'm just trying to extract some content from a large database, and then import it into my wiki.. 
[03:23:21] <Wiredtape>	:)
[03:23:26] <xtine>	ah...
[03:23:37] <Wiredtape>	I don't really care about efficiency atm..
[03:24:10] <xtine>	yeah, do what's easiest for you now
[03:24:24] <xtine>	but it's good to know if you ever run into situations where you need to speed things up.
[03:25:20] <Wiredtape>	true, hence the thank you :)
[03:26:08] <xtine>	nps
[03:26:48] <Wiredtape>	though i still can't find what the hell # means in regex..
[03:27:31] <Wiredtape>	or P for that matter..
[03:34:23] <Wiredtape>	anyone know vbulletin?
[03:40:37] <xtine>	Wiredtape: installed it, admined it, haven't hacked it before.
[03:41:31] <Wiredtape>	xtine, i'm specifically looking to find out if there's a way to view a single post by post id and if there's a way to get the raw text?
[03:41:37] <Wiredtape>	without admin rights..
[03:41:54] <Wiredtape>	any chance you know of a way? :)
[03:42:23] <xtine>	Wiredtape: you mean by like just having the url
[03:42:36] <xtine>	and then just grabbing text?
[03:42:51] <Wiredtape>	well.. it could be a simple html as well..
[03:43:04] <Wiredtape>	the closest i found was printthread.php?t=<threadnumber
[03:43:08] <xtine>	well you can view a post just by it's id
[03:43:12] <Wiredtape>	which atleast strips all of the skin..
[03:43:25] <xtine>	ah, you mean just one particular post
[03:43:29] <xtine>	not the thread associated with it
[03:43:36] <Wiredtape>	yeah..
[03:44:12] <xtine>	you can view a single post..
[03:44:46] <xtine>	like /forum/showpost?p=xxxx
[03:44:51] <xtine>	xxxxx being the number of the post
[03:44:58] <Wiredtape>	ok, let me see, one sec.. :)
[03:45:37] <xtine>	actually 
[03:45:57] <xtine>	it's showpost?p=xxx&postcount=yyy
[03:46:02] <Wiredtape>	that works, but any chance you can see the same thing in print format? 
[03:46:04] <xtine>	xxx being thread number, post number yyy
[03:46:12] <xtine>	ah
[03:46:18] <xtine>	then in that way
[03:46:22] <xtine>	you would have to make a script :)
[03:46:33] <xtine>	unless there is some extension that does that
[03:46:45] <Wiredtape>	:) well, showpost.php works fine, it's pretty simple to strip from that..
[03:46:54] <xtine>	in the source of a post
[03:46:58] <xtine>	you got a <!--- message -->
[03:47:00] <Wiredtape>	now to only make a script to get showpost.php?p=N 
[03:47:10] <xtine>	so if you are gonig to strip
[03:47:19] <Wiredtape>	i saw.. but i also need a title, which makes it only a tad more difficult.. but not by much..
[03:47:20] <xtine>	youc an easily strip out from the message comment tag
[03:47:34] <iharding>	Hi, I just upgraded and 1.11.1 broke my home rolled extensions.  No error, just redirect to main page, or "The action specified by the URL is not recognized by the wiki"
[03:47:44] <xtine>	then in php there is a function that you can strip_tags to get rid of html :)
[03:47:50] <iharding>	Any ideas?
[03:47:57] <Wiredtape>	xtine :) thx again..
[03:48:31] <Wiredtape>	iharding, explain a bit more about these extensions...
[03:48:59] <Wiredtape>	xtine, having a bit of a staying/leaving dillema?
[03:49:15] <xtine>	pressed wrong key, exited room insetad of my browser tab
[03:49:17] <xtine>	XD
[03:49:18] <iharding>	They are special page extensions using a "3 file extension" template I found.  Simple case is generating a form, creating a page based on a template.
[03:49:24] <iharding>	Works with 1.10.0
[03:49:48] <Wiredtape>	and you don't get an error but are just redirected to the main page?
[03:49:51] <iharding>	I see a caveat about "hooks" not returning a value, but I don't think I'm doing that...
[03:50:02] <Wiredtape>	that would cause a break..
[03:50:07] <iharding>	Yes, or I get the No such action message
[03:50:27] <Wiredtape>	no such action? hmm.. i've never heard of that before..
[03:50:43] <iharding>	I can paste code somewhere if you like.  I'm sure it will be obvious to a php person...
[03:50:49] <iharding>	or you can go here and see...
[03:50:51] <Wiredtape>	rafb.net
[03:51:03] <Wiredtape>	and the link please..
[03:51:15] <iharding>	http://gmpartswiki.com and click on the My Part List link
[03:51:32] <iharding>	The other one is just listed under special pages now, called Create Part.
[03:53:31] <Wiredtape>	iharding, the problem is either $wgarticlepath in localsettings or your htaccess file
[03:53:45] <Wiredtape>	take a look at http://gmpartswiki.com/w/Special:PartList
[03:54:18] <Wiredtape>	whereas the link on the left links to /wiki/index.php/Special:PartList/
[03:54:51] <shortcircuit>	Does anybody else watch their wiki's Recent Changes RSS/Atom feed using Google Reader?
[03:55:45] <iharding>	http://rafb.net/p/VkfgYU62.html
[03:55:52] <shortcircuit>	I've noticed a problem where changes show up multiple times in Reader's feed history.  I just wondered if that happened to anybody else.
[03:56:27] <shortcircuit>	It doesn't seem to happen to any of the other feeds I watch, just my wiki feed.
[03:57:05] <iharding>	Code is for this page
[03:57:10] <iharding>	http://gmpartswiki.com/w/Special:CreatePart
[03:57:36] <Wiredtape>	iharding, the problem isn't in the extensions code
[03:57:37] <iharding>	That one actually creates the form, but the action is not found.  Gah
[03:58:29] <iharding>	I'm using the same LocalSettings though, and all else seems to work... 
[03:58:54] <Wiredtape>	:) as i said check your articlepath and htaccess..
[04:01:14] <iharding>	$wgArticlePath      = "/w/$1";
[04:01:30] <Wiredtape>	and script path?
[04:01:41] <iharding>	  Alias /w /usr/pkg/share/httpd/htdocs/wiki/index.php
[04:01:41] <iharding>	  Alias /index.php /usr/pkg/share/httpd/htdocs/wiki/index.php
[04:02:39] <Wiredtape>	iharding, ok what creates the partlist portlet?
[04:02:51] <Wiredtape>	did you put it directly into the skin?
[04:03:22] <iharding>	I put that link in the skin, but that's a red herring.  The Create Part extension shows the problem
[04:03:41] <iharding>	http://gmpartswiki.com/w/Special:CreatePart
[04:04:14] <iharding>	The form gets generated, but any submissions go to the search page.  That's not supposed to happen, and didn't with 1.10
[04:04:27] <iharding>	I think I missed a return statement or something, and 1.10 was more forgiving.
[04:05:50] <Wiredtape>	iharding, i'm pretty sure this has more to do with proper linking than with a missing return statement.. though I can't tell for sure, the only thing I can tell is that if I click on the part links on your sidebar the link points to an invalid location: /wiki/index.php/Special:.../ 
[04:08:12] <ehn>	a mediawiki installation is slightly less usable out of the box than wikipedia, because it's lacking a lot of useful templates. is there some way to get commonly useful templates, such as ambox, as opposed to the ones which are specific for an online encyclopedia, such as most of the infoboxes, apart from manually copying them (and all templates they depend on, css code and images) from wikipedia?
[04:08:36] <ehn>	sorry if this is a common request. i wasn't able to find anything about it in the faqs.
[04:09:04] <Wiredtape>	ehn, i was actually thinking of something like this.. but the only option i can think of is a nice list of useful templates and then using special:export
[04:09:07] <jeremyb>	ehn: there's a template sharing project to get templates sync'd within WMF but it's stalled last i checked.
[04:09:29] <jeremyb>	[[WP:TSP]] or something
[04:09:36] <ehn>	checking...
[04:10:19] <jeremyb>	yeah, it's pretty much the same shape from last i checked
[04:10:35] <jeremyb>	meta sends you to en.wp which sends you to meta
[04:11:54] <jeremyb>	respective talk pages are similarly inactive
[04:12:05] <Splarka>	would probably also have to be an extension sync and tidy sync
[04:12:18] <jeremyb>	erm?
[04:13:13] <Splarka>	many templates on wikimedia won't work without Tidy enabled (if the html tags are split via transclusion or parserfunction)
[04:13:38] <Splarka>	<table><tr><td>foo</td>{{#if:1|<td>bar</td>}}</tr></table>
[04:13:46] <ehn>	playing around with special:export/import for now... :/
[04:13:47] <Splarka>	and many won't work without the proper extensions enabled/installed
[04:13:59] <jeremyb>	ohhh, didn't get tidy was an ext
[04:14:07] <jeremyb>	so what's "extension sync" then?
[04:14:08] <Splarka>	it isn't
[04:14:11] <CIA-6>	03laner * r31135 10/trunk/extensions/SmoothGallery/ (4 files): 
[04:14:11] <CIA-6>	* Refactored code
[04:14:11] <CIA-6>	** Broke apart parsing and rendering
[04:14:11] <CIA-6>	** Moved most functions into classes
[04:14:11] <CIA-6>	** Fixed broken gallery checks
[04:14:12] <CIA-6>	** Removed unneeded global $wgSmoothGalleryArguments
[04:14:18] <jeremyb>	CIA-6: quit that!
[04:14:23] <Wiredtape>	tidy is a localsettings option
[04:14:26] <Splarka>	but it isn't enabled by default IIRC (for example, Wikia doesn't have it)
[04:14:30] <Ryan_Lane>	jeremyb: thats a bot ;)
[04:14:35] <jeremyb>	i know!
[04:14:39] <Wiredtape>	lol
[04:14:55] <jeremyb>	Splarka: so why is it on anywhere?  just for the templates?
[04:15:14] <Splarka>	cleaner cleanup of messy wikicode/html I guess
[04:15:21] <Splarka>	but it conflicts with some things and is buggy
[04:15:31] <jeremyb>	does it improve UA compat?
[04:15:41] <Ryan_Lane>	Splarka: i'm sure it cleans up my messy html
[04:16:06] <Splarka>	a broken table inside a table in wikicode renders amazingly bad, column-one ends up inside the bodycontent
[04:16:09] <jeremyb>	i don't see how the example template you gave would be a problem anyway
[04:16:29] <Splarka>	<table><tr><td>foo</td>{{#if:1|<td>bar</td>}}</tr></table> won't work without tidy
[04:16:42] <Splarka>	the sanitizer escapes the inner TD to &lt;td&gt;
[04:17:09] <jeremyb>	sanitizer?
[04:17:34] <Splarka>	http://svn.wikimedia.org/viewvc/mediawiki/trunk/phase3/includes/Sanitizer.php
[04:19:34] <Wiredtape>	i must say that there is too much project related lingo involved :) i have no idea how a new mw user can learn all this stuff...
[04:19:38] <Splarka>	https://bugzilla.wikimedia.org/show_bug.cgi?id=9252  <-- here was a bug with tidy
[04:20:13] <Splarka>	or rather, a bug in mediawiki only visible when tidy was enabled
[04:21:31] <Wiredtape>	well g'night!
[04:21:34] <Splarka>	it is very annoying, MediaWiki should fork tidy and make it core and merged with the sanitizer (and not optional), rather than try to support two types of template syntax forever
[04:22:14] *Splarka 	grumps
[04:22:35] *jeremyb 	grumps
[04:22:56] <Splarka>	harrumph!
[04:23:35] <Splarka>	anyway, jeremyb, at least the CIA here reports naughty things immediately, instead of 25 years later
[04:24:01] <jeremyb>	naughty?
[04:24:02] <Splarka>	<CIA-6> Oh, we waterboarded some prisoners and tried to assassinate castro
[04:24:16] <jeremyb>	and why are you so much faster than everyone else?
[04:24:34] <Splarka>	that's what all the girls ask me
[04:24:36] <iharding>	Here's the prob... http://www.gossamer-threads.com/lists/wiki/mediawiki/104782
[04:24:45] <jeremyb>	hardy har har
[04:31:29] *jeremyb 	wonders how wmf demographics compare with wikia
[04:31:35] <jeremyb>	specifically geography
[04:34:46] <Splarka>	wikia keeps detailed stats via google analytics...
[04:34:59] <jeremyb>	hosted?
[04:35:05] <Splarka>	private
[04:35:18] <Splarka>	ad revenue and all
[04:35:19] <jeremyb>	erm... hosted? :P
[04:35:37] <jeremyb>	i.e. on google's servers or on wikia servers?
[04:35:49] <Splarka>	the stats? on google's
[04:35:57] <jeremyb>	so hosted
[04:36:09] <Splarka>	well, not so mcuh
[04:36:13] <Splarka>	^much
[04:36:17] <Splarka>	more like, generated
[04:36:34] <jeremyb>	mozilla has outgrown the internal version of a few analytics solutions and is now moving to google hosted
[04:36:52] <jeremyb>	(was urchin)
[04:36:54] <Splarka>	they add the GA js to each page view, users view a page and their browser loads the remote .js which google sees and records their visit
[04:37:26] <jeremyb>	i think it's more complicated than that
[04:37:29] <Splarka>	and then the wikia staff with access to that GA account can click and view the infos... would you call that hosted?
[04:37:41] <jeremyb>	it queries all kinds of info about the browser via js
[04:37:52] <jeremyb>	yes i would
[04:38:02] <Splarka>	I'd call it "offsite"
[04:38:44] <Splarka>	anyway, I wrote a temp JS web-bug for Wikia to log activity like action=, namespace, anon/logged, time, and browser type (before they switched to GA), but no geographical info was ever taken
[04:38:55] <Splarka>	as for Wikimedia...
[04:39:29] <Splarka>	the toolserver does do the localization watchlist message... but I don't know if that data is saved anywhere
[04:39:52] <jeremyb>	localization?
[04:40:07] <jeremyb>	you know there's a toolserver project just for tracking page stats right/
[04:40:08] <Splarka>	 if (wgPageName == "Special:Watchlist")
[04:40:08] <Splarka>	     addOnloadHook((function (){document.write('<script type="text/javascript" src="http://tools.wikimedia.de/~gmaxwell/cgi-bin/geonotice.py"><\/script>')}));
[04:40:09] <jeremyb>	?*
[04:40:24] <jeremyb>	oh, geonotice
[04:40:30] <Splarka>	right, localized so you don
[04:40:36] <Splarka>	't spam new yorkers with a SF meetup
[04:40:40] <jeremyb>	in my head localization==language
[04:40:54] <Splarka>	well, geo, localization..... geolocation
[04:41:05] <jeremyb>	geonotice*
[04:41:10] <jeremyb>	!
[04:41:16] <Splarka>	localized geonotices
[04:42:47] <Splarka>	maybe if you're feeling lucky you could ask domas to log the country a the squid level (via checking the IP and using a geolocation table, and then discarding the IP)
[04:43:50] <Splarka>	although the logs would probably be better kept separate from the page access logs, for privacy
[04:44:24] <Splarka>	maybe something like an hourly breakdown of how many visits each country made to each project on each subdomain
[04:46:33] <jeremyb>	hrmmm... i think i'd want breakdown of editors vs. readers, logged in vs. anon, mainspace vs. all else, etc.
[04:46:37] *jeremyb 	dreams
[04:46:44] <Splarka>	well, sure, who wouldn't
[04:47:16] <Splarka>	I did it because I was trying to convince Wikia to: remove ads from edit/history/diff pages, remove ads from non-main namespaces, remove ads for logged in users, etc
[04:47:26] <Splarka>	since they made up, total, less than 50% of all accesses
[04:47:52] <Splarka>	mainly because the ads made the edit box smaller
[04:53:10] <Werdna>	TimStarling: around? I want to ask you a bit about the category metadata stuff you were up to last night.
[05:06:37] <Nineth>	I am trying to upload a txt file but getting error - ".txt" is an unwanted file type . I added $wgFileExtensions[] = 'txt' on localsettings.php but no luck. How can i enable uploading of .txt files?
[05:08:24] <Nineth>	Below that error message it displays - List of allowed file types: png, gif, jpg, jpeg, txt 
[05:10:39] <Nineth>	Then I found a bug ;) - The file I was trying to upload was phpwatermark.php.txt. I renamed to phpwatermark.ph_.txt and it works now 
[05:15:33] <wikibugs>	03(NEW) File uploading - filename issue - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13088 minor; normal; MediaWiki: Uploading; (praveen)
[05:15:48] <wikibugs>	03(mod) Though user is logged in, switching pages seems to force a logout - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13082  +comment (10cwisniewski)
[05:16:21] <Werdna>	hey brion 
[05:16:37] <brion>	oy
[05:17:36] <linuxpoet>	what is the schema for page_props?
[05:17:46] <linuxpoet>	I am getting an error on trunk
[05:19:13] <wikibugs>	14(DUP) File uploading - filename issue - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13088  +comment (10brion)
[05:19:15] <wikibugs>	03(mod) blacklist extension handling is broken - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=11142  +comment (10brion)
[05:19:26] <Charitwo>	o_O
[05:19:34] <Charitwo>	oh, just brion poking
[05:19:35] <brion>	i seem to remember that one having been fixed since last release, but might misremember
[05:19:37] <brion>	anyone recall?
[05:19:52] <Charitwo>	not off the top of my head
[05:21:53] <TimStarling>	I don't remember seeing that bug before
[05:22:19] <linuxpoet>	ok what is the easiest way to make it so only logged in people can edit?
[05:22:36] <linuxpoet>	I found page_props in archives and edited the sql to make it work with postgresql
[05:22:39] <wikibugs>	03(mod) Though user is logged in, switching pages seems to force a logout - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13082  (10overlordq)
[05:23:06] <wikibugs>	03(FIXED) bug on r31113 - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13085  +comment (10tstarling)
[05:23:43] <TimStarling>	I've been doing a bit of work on usability stuff
[05:24:09] <TimStarling>	I fixed the stupid maintenance categories cluttering up the bottom of the page yesterday
[05:24:14] <TimStarling>	community willing
[05:24:29] <TimStarling>	today I will commit a fix for red links that tell you you're blocked
[05:24:59] <Charitwo>	before they try to edit a page?
[05:25:31] <TimStarling>	a lot of readers don't know what a red link means
[05:25:47] <TimStarling>	so sometimes, the click one, and find that they're part of a range block, or their ISP proxy is blocked
[05:25:51] <TimStarling>	and they think they're blocked from reading
[05:26:21] <Charitwo>	ah, that'll be handy
[05:26:24] <TimStarling>	it's just stupid, my patch will make it so that they're redirected to action=view if they're not allowed to edit
[05:26:43] <TimStarling>	then when they click the edit tab, they'll have a higher chance of knowing what they're blocked for
[05:26:50] <linuxpoet>	anyone?
[05:27:18] <TimStarling>	linuxpoet: have you tried the FAQ?
[05:27:26] <linuxpoet>	??faq
[05:27:33] <brion>	hm
[05:27:49] <brion>	could work
[05:28:17] <TimStarling>	it works by having red links point to action=editrl
[05:28:27] <TimStarling>	name chosen so it still looks like an edit link on mouseover
[05:28:51] <TimStarling>	we already have two actions that point to the edit page, edit and submit, there's no reason why we can't add a third
[05:33:28] <linuxpoet>	sheesh this is a new world
[05:35:20] <Splarka>	linuxpoet: http://www.mediawiki.org/wiki/Help:User_rights
[05:35:37] <Splarka>	tim: will the wgAction JS variable be 'edit' or 'editrl' ?
[05:36:08] <Splarka>	or will it 302 to action=edit if not blocked, and action=view if blocked?
[05:38:26] <linuxpoet>	Splarka: Yeah I found it, thanks :) $wgGroupPermissions['*']['edit'] = false;
[05:39:21] <brion>	maybe editlink instead of editrl?
[05:39:30] <brion>	for legibility's sake :)
[05:39:33] <TimStarling>	Splarka: presumably it will be editrl
[05:39:59] <TimStarling>	brion: that's a nice pronounceable name, there's just the slight problem that it doesn't mean anything
[05:40:11] <TimStarling>	but meaning comes from use, right?
[05:40:38] <brion>	editrl doesn't mean anything either, and it's unpronouncable
[05:40:44] <Splarka>	how about editview, which will 302 to view if the user cannot edit that page for any reason (user is blocked and not editing talk page, page protected, page in protected namespace or user css/js subpage, $wgGroupPermissions['*']['edit'] = false;, etc)
[05:40:51] <Pathoschild>	TimStarling: I'm just wondering, why not just implement this on the edit form? Then you don't need to add a new action, which needs to be accounted for in scripts and such. Would that be more difficult to do?
[05:40:54] <Splarka>	action=editview ^_^
[05:40:54] <TimStarling>	RL stands for something that may mean something to somebody
[05:41:00] <TimStarling>	e.g. me
[05:41:12] *Splarka 	hides
[05:41:36] <brion>	action=create ?
[05:42:03] <brion>	or just drop the 'action' entirely, and stick an edit form on the view page for them redlinks
[05:42:09] <TimStarling>	action=create is nice, except that the whole point is to distinguish between edit tab and red link
[05:42:10] <Splarka>	red-edit, rededit, reedit, redit. createedit, creadit, credit...
[05:42:11] <brion>	i'm just tossing out ideas here
[05:43:46] <TimStarling>	action=view for nonexistent articles is fairly nice-looking on wikipedia at least
[05:43:57] <TimStarling>	it would seem like a shame to me to mess it up with an edit form
[05:44:07] <TimStarling>	not to mention the fact that that edit form takes about a second to render
[05:44:15] <TimStarling>	PHP... is not fast :)
[05:44:42] <Charitwo>	using view is reasonable enough, without needing to add a new action
[05:45:02] <TimStarling>	anyway, you've all had 5 years to come up with crazy ideas for fixing this problem
[05:45:11] <TimStarling>	I don't know why you all want to change it now that it's actually written
[05:46:05] <TimStarling>	surely I get some amount of design leeway as the guy who managed to get off his arse and write it?
[05:47:32] <Splarka>	just sayin, if action=editrl would 302 to action=edit if the user *could* edit, it would make userscripts that much less confusing ^_^
[05:47:47] <Splarka>	of course, that runs the risk of turning mediawiki's 302 redirects into a programming language
[05:47:52] <Charitwo>	just ideas, do what you want lol
[05:49:05] <TimStarling>	Splarka: "that much less" being how much exactly?
[05:49:23] <TimStarling>	(action=="edit" || action=="editrl") instead of (action=="edit") ?
[05:49:44] <Splarka>	don't forget action="submit"
[05:50:06] <Splarka>	so, only 50% more "complex"
[05:51:11] <Splarka>	(and of course, some will drool that the opportunity to distinguish between a tab and redlink click)
[05:53:34] <Werdna>	TimStarling: did you read what I wrote on wikitech-l about category counts?
[05:53:38] *Splarka 	casts his vote for "perfect as is" and gives Tim a cookie
[05:53:58] <TimStarling>	Werdna: I may have done
[05:54:10] <TimStarling>	although I was kind of annoyed at Magnus at the time so I wasn't paying much attention
[05:54:54] <Werdna>	TimStarling: well, Simetrical suggested a category table, which I think could be used to store your "don't show this category on pages" metadata, etc.
[05:55:35] <Werdna>	and the category table could also include the number of items in that category, and other goodies.
[05:56:05] <TimStarling>	page_props isn't limited to categories
[05:56:39] <Werdna>	ah
[05:56:51] <TimStarling>	it could be used for image overlays as well, that's the most important potential application
[05:57:07] <Werdna>	right, I heard you talking about that last night.
[05:57:09] <TimStarling>	the ability to overlay an image with a warning if it's listed on IfD or something
[05:57:46] <Werdna>	do you think that a 'category' table is a good spot to store category membership-count data?
[05:57:56] <TimStarling>	the big unknown with category counting is lock contention
[05:58:18] <Werdna>	i.e. Users x,y both add [[Category:Bad stuff]] to articles A and B at the same time?
[05:58:19] <TimStarling>	hard to know how the site would react to such a thing
[05:58:32] <TimStarling>	I guess if you kept the transaction very, very short, it wouldn't be a problem
[05:59:07] <TimStarling>	well yeah, but the problem becomes severe if 500 users are all trying to edit [[Category:Bad stuff]]
[05:59:17] <TimStarling>	and the locks pile up on top of each other like a traffic jam
[05:59:27] <TimStarling>	until the thread count of the server is exhausted and the site goes down
[05:59:34] <Werdna>	Well, if you write a query like UPDATE category set cat_count=cat_count+1 where cat_id=568795, then, if each one is executed in sequence, then there's no unexpected behaviour.
[05:59:52] <TimStarling>	it's not a consistency problem, it's a performance problem
[06:01:01] <Werdna>	you mean that repeatedly changing a particular row in a table causes problems because the transactions can't all be handled at once?
[06:02:14] <TimStarling>	when you run a query like that UPDATE query, it obtains a lock on the row
[06:02:26] <TimStarling>	and that lock is held until the transaction is committed
[06:03:01] <Werdna>	right. I hadn't considered that (and I don't really fully understand it)
[06:03:06] <TimStarling>	if another process attempts to run an UPDATE query on the same row, it will wait until the COMMIT of the first process
[06:03:27] <Werdna>	what determines how long a transaction takes to commit? Is there a $dbw->commit() function that should be called right after updating the row?
[06:04:09] <TimStarling>	yes, you can call $dbw->commit(), and I suppose you'd have to
[06:04:53] <TimStarling>	it shouldn't be a problem if you do, presumably the lock time would only be a few milliseconds, so you'd be able to run tens or hundreds of such queries per second
[06:05:01] <Werdna>	per category
[06:05:05] <TimStarling>	right
[06:05:20] <TimStarling>	it'd only be a problem if you held it for several seconds
[06:05:28] <Werdna>	i.e. for an entire save page-load
[06:05:35] <TimStarling>	yes
[06:05:52] <TimStarling>	the problem with using more transactions is that you can lose consistency if apache crashes
[06:06:07] <TimStarling>	but we already have that problem and this won't make it significantly worse
[06:06:34] <TimStarling>	e.g. you can block someone and have it not show up in the block log
[06:06:51] <OvrLrd-Q>	couldn't you just throw in a 'these numbers may be approximate' and just queue up the category stuff to only run every $x so you dont have to lock on every update?
[06:06:56] <TimStarling>	because the logging is done in a separate transaction, and apache can crash before it's committed
[06:07:35] <Werdna>	Well, you could hold the lock only for the time that you're updating categorylinks, if you're worried about consistency
[06:08:04] <TimStarling>	this is a different sort of consistency...
[06:08:49] <Werdna>	what kind, then?
[06:08:57] <TimStarling>	transactions are meant to be used to preserve consistency, so that if one query in a group of queries fails, you can roll back the transaction and not have the inconsistent scenario of a half-done operation
[06:09:12] <Werdna>	right..
[06:09:14] <TimStarling>	a half-done operation like an unlogged block
[06:09:33] <TimStarling>	so from that point of view, it's better to have a few long transactions
[06:09:44] <TimStarling>	but from a performance point of view, it's better to have short transactions
[06:09:58] <Werdna>	or a category which has been recorded, but which has not been added to the total category membership count
[06:10:20] <TimStarling>	exactly, and that's entirely possible
[06:10:41] <TimStarling>	the category count would indeed drift due to random failures
[06:10:56] <Werdna>	so, as I said, what's wrong with treating "add the category member to categorylinks, and then add to the category count" as a transaction?
[06:11:42] <TimStarling>	may be possible, yes
[06:12:10] <Werdna>	If that transaction fails - big deal, we can add it next time the page is null-edited.
[06:16:01] <TimStarling>	you know, the inconsistency that Magnus pointed out could be solved just by removing the category count from category pages with more than 200 members
[06:16:30] <Pathoschild>	Or reword to "There are 200 members shown below (view next 200)."?
[06:16:49] <Pathoschild>	Or (view next page).
[06:16:54] <TimStarling>	Pathoschild: yes, but that's redundant
[06:17:07] <TimStarling>	yeah, with "next page" it's not redundant
[06:17:14] <TimStarling>	but obviously "next 200" implies that you're seeing 200 already
[06:17:22] *Pathoschild 	nods.
[06:17:23] <MrZ-cat>	TimStarling: that would kind of be dodging the real issue though
[06:17:25] <MrZ-cat>	;)
[06:17:34] <TimStarling>	anyway, just because someone put a category count on the page doesn't mean it's actually useful information
[06:17:45] *MrZ-cat 	doesn't use them anyway
[06:17:47] <Werdna>	TimStarling: sure, but I've been asked about actually counting items in a category pretty regularly.
[06:18:25] <MrZ-cat>	TimStarling: actually, except for categories with tons of members, its not all that helpful
[06:18:41] <MrZ-cat>	and ones like that wouldn't have an accurate count...
[06:19:03] <TimStarling>	MrZ-cat: well, Werdna is proposing a way to accurately count huge categories
[06:19:09] <MrZ-cat>	ah
[06:19:17] *MrZ-cat 	scrolls up
[06:20:40] <MZMcBride>	couldn't action=create help with bugs like this? https://bugzilla.wikimedia.org/show_bug.cgi?id=10730
[06:22:23] <wikibugs>	03(mod) Request layout change on editing buttons - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13065  (103tx38k)
[06:22:41] <TimStarling>	no
[06:23:15] <MZMcBride>	is there any efficient way to generate a list of pages a user has created?
[06:23:32] <TimStarling>	yes, a number of them
[06:23:46] <MZMcBride>	the current toolserver tools all die on en.wiki
[06:24:12] <MZMcBride>	any hints?
[06:24:38] <TimStarling>	EditPage/Article are already well aware of whether the page exists or not, so there's no need for action=create
[06:25:17] <TimStarling>	they all involve schema changes, so probably not much help for the toolserver
[06:25:39] <MZMcBride>	ok, thanks
[06:28:50] <TimStarling>	people always want more indexes, don't they?
[06:28:56] <TimStarling>	no matter how many you give them
[06:29:21] <TimStarling>	what we need is an implementation of Grover's algorithm for MySQL
[06:30:10] <MZMcBride>	well, a mountain of data is never helpful to anyone; people will always want stats and useable info from the data
[06:36:36] <wikibugs>	03(mod) Request layout change on editing buttons - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=13065  (103tx38k)
[06:37:15] <TimStarling>	no takers?
[06:37:18] *Werdna 	googles Grover's algorithm
[06:37:55] <Werdna>	a quantum algorithm for searching in O(N^1/2)?
[06:40:21] <TimStarling>	yeah, then we won't have to worry about indexing anymore, we can just let people search the database linearly
[06:40:53] <Werdna>	TimStarling: fair enough. I will require the Wikimedia server farm to be replaced with quantum computers, and then I'll think about it.
[06:42:03] <Werdna>	TimStarling: we might need another fundraising drive for this one
[06:42:32] <TimStarling>	DARPA thinks they can build one for $100M, but it would only have 100 qubits
[06:42:38] <TimStarling>	our database is, what, 100GB?
[06:43:10] <TimStarling>	so a million dollars per qubit, need about a trillion of them...
[06:43:29] <TimStarling>	hmmm, I think that's more than the GDP of the earth
[06:43:34] <Splarka>	does wmf accept donations from parallel universes?
[06:43:35] <TimStarling>	so donation drive might not cut it
[06:44:04] <TimStarling>	Splarka: if not, it damn well should
[06:44:12] <TimStarling>	time for some real work
[06:45:13] <Werdna>	TimStarling: slightly back on track, should I have a crack at the category counting?
[06:45:38] <TimStarling>	if you like
[06:45:58] <TimStarling>	I'd put it in LinksUpdate.php if I were you, that's the easiest way to share a transaction with categorylinks
[06:46:21] <TimStarling>	plus the category change information is there and available
[06:46:54] <MZMcBride>	will __HIDDENCAT__ allow the talk page to be categorized instead of the subject page?
[06:46:56] <Werdna>	that was my intention.
[06:47:46] <TimStarling>	MZMcBride: you mean will the category show up at the bottom of the talk page?
[06:47:53] <MZMcBride>	yeah
[06:47:58] <TimStarling>	well, that would be easy enough to implement
[06:48:00] <TimStarling>	but it's not
[06:48:44] <TimStarling>	but you still have the big ugly box at the top of the article to remind you of what category it's in
[06:48:47] <MZMcBride>	that's been the big complain I've heard; that you don't want articles to be classified in "Articles needing cleanup", but their talk pages should be
[06:48:57] <MZMcBride>	right
[06:49:37] <TimStarling>	anyway, if people wanted the talk pages categorised, they'd put category tags on them
[06:49:41] <TimStarling>	right?
[06:50:04] <TimStarling>	or do they want the talk page categorised, but also the big ugly box on the article?
[06:50:04] <Werdna>	yeah, but the categories wouldn't show up
[06:50:08] <TimStarling>	and they don't want to edit both?
[06:50:14] <MZMcBride>	the latter
[06:50:45] <MZMcBride>	box shows up to a standard reader indicating an issue with the article, but the reader isn't forced to see a bunch of categories at the bottom that they probably don't need
[06:50:45] <TimStarling>	do you think the feature as implemented will be controversial?
[06:51:31] <Werdna>	TimStarling: don't ask me, we all know my track record on judging anticipated controversy.
[06:51:37] <TimStarling>	one simple thing to do would be to collapse the hidden categories with javascript, instead of removing them completely
[06:51:39] <MZMcBride>	well, currently it just hides the category from the bottom of pages if the Category: page contains __HIDDENCAT__ is present, right?
[06:51:44] <TimStarling>	yes
[06:52:01] <MZMcBride>	no, that won't be controversial
[06:52:13] <MZMcBride>	and for certain pages like the Main Page, it would allow us to remove CSS hacks
[06:52:40] <TimStarling>	good
[06:52:42] <Splarka>	how about __TALKCAT__ that only does one thing: adds a link to the subject *and* talk page in the category view (with the categorized page bolded)
[06:53:13] <Splarka>	so [[Foo]] with the cat would show as '''Foo''' (talk).... and Talk:Bar would show as Foo ('''Talk''')
[06:53:50] <TimStarling>	sure
[06:53:58] <TimStarling>	although maybe a bit more thought needs to go into the naming
[06:54:27] <Splarka>	__MAGICALDEVICETODOWHATIWANTTOCATEGORIESONTALKPAGESBOYTHISISALONGMAGICEORD__
[06:56:33] *siebrand 	blinks.
[06:57:41] *Splarka 	likes to make random naming interjections
[06:59:38] <Charitwo>	s/eord/word
[07:01:09] <Werdna>	Charitwo: that regex wouldn't do anything
[07:01:29] <Charitwo>	Werdna: it's not a regex
[07:01:38] <Charitwo>	it's a typo fix
[07:01:44] <Splarka>	/i
[07:01:50] <Werdna>	:)
[07:01:58] <Werdna>	Splarka got it
[07:02:07] <Werdna>	Charitwo: s/x/y/ is regex syntax
[07:02:22] <Splarka>	except you'd get ...YTHISISALONGMAGICword__
[07:05:07] <Charitwo>	ah well,  {"ra", "r"}{"r", "ar"}
[07:07:25] <TimStarling>	according to the PHP manual, list() doesn't return anything
[07:07:37] <TimStarling>	but according to the another part of the manual, you can use it in a while condition
[07:07:50] <TimStarling>	while (list($key, $val) = each($fruit)) {
[07:08:01] *Werdna 	shrugs. Try it.
[07:08:08] <TimStarling>	I have, it turns out, it does actually return something
[07:08:40] <Werdna>	ah
[07:08:46] *Werdna 	notes that this is why we have wikis
[07:08:50] <TimStarling>	it returns whatever was assigned to it
[07:09:12] <TimStarling>	or maybe in PHP, an assignment always returns the RHS
[07:09:55] <Werdna>	you mean ($b = $a) == $a?
[07:10:12] <TimStarling>	> var_dump( list($a) )
[07:10:23] <TimStarling>	PHP Parse error:  syntax error, unexpected ')', expecting '=' in...
[07:10:47] <TimStarling>	all the smarts are in the compiler in PHP
[07:11:12] <Werdna>	well, php was intended to be perlish, wasn't it?
[07:11:20] <TimStarling>	perl has operators
[07:11:21] <Werdna>	and that's what perl is all about - special cases galore.
[07:11:37] <TimStarling>	this is a special case syntax structure, in perl, everything's an operator
[07:12:20] <TimStarling>	in perl, you can have a list on the LHS, because perl has a very generic way of handling assignments to lists
[07:12:28] <TimStarling>	in PHP, it's a compiler hack
[07:13:23] <Werdna>	hmm.. interesting.
[07:13:35] <OvrLrd-Q>	mmmmm Perl
[07:14:25] <OvrLrd-Q>	so you can'd do something like ($foo, $bar, $baz, $quux) = RHS?
[07:14:29] <OvrLrd-Q>	*can't
[07:14:38] <Werdna>	not in PHP
[07:14:47] <OvrLrd-Q>	well that's horribly evil
[07:15:01] <TimStarling>	well, that's not a list in PHP
[07:15:04] <Werdna>	you use list($foo, $bar, $baz, $quux) = RHS
[07:15:08] <TimStarling>	so the example isn't exactly valid
[07:15:37] <OvrLrd-Q>	oh
[07:15:55] <OvrLrd-Q>	still the devil language I say ;)
[07:16:55] <TimStarling>	this is a better demonstration of the limitations of PHP:
[07:17:02] <TimStarling>	> print some_function()[0];
[07:17:03] <TimStarling>	PHP Parse error:  syntax error, unexpected '[' in
[07:17:36] <TimStarling>	because square brackets can only follow variables, not expressions
[07:17:55] <TimStarling>	> print ($a)[0]
[07:17:55] <TimStarling>	PHP Parse error:  syntax error, unexpected '['
[07:18:07] <Werdna>	this is why I like perl better
[07:18:12] <Werdna>	it's much more logical.
[07:19:05] <OvrLrd-Q>	throw the schwartzian transform at a PHP coder probably make his head 'splode :D
[07:20:06] <Werdna>	jeez, I'm way behind on this channel
[07:20:15] <Werdna>	that's the second time I've seen the name of an algorithm and googled it today
[07:31:58] <TimStarling>	quick quiz:
[07:32:03] <TimStarling>	> print true || false && false
[07:32:16] <TimStarling>	is it true or false?
[07:32:44] <AaronSchulz>	I should do my context-free grammar lab now
[07:33:13] <csaba>	Is there a way to create a page which redirects to a website that is outside the wiki?
[07:33:21] <csaba>	Like #REDIRECT [http://www.google.com]
[07:34:27] <Werdna>	TimStarling: false. It travels left to right, I think.
[07:34:57] <TimStarling>	nope, it's true
[07:35:05] <Werdna>	...
[07:35:06] <TimStarling>	and it's relevant because someone got this quiz wrong in EditPage.php
[07:35:15] <Werdna>	I always use brackets anyway.
[07:35:21] <TimStarling>	&& has higher precedence than ||
[07:35:32] <TimStarling>	this is true in C and C-like languages, and in discrete mathematics
[07:35:44] <TimStarling>	and it's true in PHP as well
[07:35:58] <csaba>	and probably every other prog language
[07:36:13] <TimStarling>	( true || false && false ) == ( true || ( false && false ) )
[07:36:19] <csaba>	where can i edit the list of allowed html tags?
[07:36:25] <Werdna>	csaba: Sanitizer.php
[07:36:29] <csaba>	thanks
[07:37:18] <AaronSchulz>	TimStarling: I want to see the re-write rules ;)
[07:38:07] <TimStarling>	Werdna: http://svn.wikimedia.org/viewvc/mediawiki/trunk/phase3/includes/EditPage.php?view=diff&r1=24561&r2=24562
[07:38:17] <TimStarling>	line 327 on the RHS
[07:39:02] <TimStarling>	look familiar?
[07:39:43] <AaronSchulz>	naughty
[07:40:23] <Werdna>	ah
[07:40:54] *Werdna 	should be more careful adapting code.
[07:41:20] <TimStarling>	don't edit it now, you'll conflict with me
[07:41:29] <Werdna>	I know, I figured as much
[07:52:21] <Nikerabbit>	hello
[07:53:25] <Splarka>	csaba: http://www.mediawiki.org/wiki/Manual:$wgDisableHardRedirects  per your other question (sorta)
[07:54:12] <csaba>	thanks i fixed it by enabling iframes.. the point is the external page can be seen :)
[07:56:23] <AaronSchulz>	Werdna: I'd laugh at you if I didn't add funky array += code for a while ;)
[07:58:33] <Werdna>	now now
[07:58:34] <Werdna>	be nice
[07:59:15] *Werdna 	is off dividing polynomials
[07:59:56] <CIA-6>	03siebrand * r31136 10/trunk/extensions/GoogleMaps/ (17 files): Import as delivered by Evan Miller <emmiller AT gmail DIT com>
[08:00:17] <Sasa^Stefanovic>	DIT ? :)
[08:00:26] <Werdna>	?
[08:00:55] <MZMcBride>	really trying to confuse the spambots, eh?
[08:00:57] <Splarka>	http://en.wiktionary.org/wiki/dit
[08:01:16] <Splarka>	dontcha speak morse?
[08:02:42] <Splarka>	dit dit dit space dit dit space dit dah dit dit space dit dah dit dit space dah dit dah dah space space dit dit dit space dit dah space dit dit dit space dit dah space space dah dit dit space dah dah dah space dah space dit dit dit space space dit dah space dit dah dit space dit space space dit dit dah dit space dah dah dah space dit dah dit space space dah dit dah space dit dit space dah dit dit space dit dit dit
[08:02:48] *Werdna 	stabs Splarka 
[08:02:53] <MZMcBride>	heh
[08:03:05] <Splarka>	(translation: silly sasa dots are for kids)
[08:03:32] *Splarka 	bleeds on Werdna
[08:05:51] <Splarka>	mmm, that's odd
[08:06:14] <Splarka>	http://test.wikipedia.org/w/index.php?title=800K&curid=13812&action=history  799,999 on the history (shoulda been 800,000 except it stripped the last newline?) but...
[08:06:53] <Splarka>	http://test.wikipedia.org/wiki/Special:Recentchanges  on recent changes, I added +700,000 to +999,996 + 3... and I just realized the math works out, nevermind
[08:07:11] <OvrLrd-Q>	lol
[08:07:18] <Splarka>	(part of the, realize the answer after pasing the URL that I regularly go through)
[08:09:50] <OvrLrd-Q>	silly noarticletext div
[08:10:54] *OvrLrd-Q 	moves the article to 799K ;)
[08:11:17] <Splarka>	OQ: purge the relevant mediawiki message to get rid of it
[08:11:25] <OvrLrd-Q>	?
[08:11:34] <Splarka>	the div
[08:12:12] <OvrLrd-Q>	it's in the Log in/create account link at the top
[08:12:27] *OvrLrd-Q 	has no clue which message that is :)
[08:13:05] <TimStarling>	why do we ignore some permissions errors on preview, but not others?
[08:13:33] <Splarka>	OvrLrd-Q: purged, try reloading
[08:13:45] *Splarka 	missed that one since he was logged in
[08:14:05] <OvrLrd-Q>	works now
[08:14:31] <Splarka>	it was MediaWiki:Userlogin?action=purge
[08:14:47] <MZMcBride>	Tim: because developers are silly, unpredictable creatures?
[08:15:09] <OvrLrd-Q>	Ahhhh
[08:16:19] <Werdna>	TimStarling: We should prooobably have a big ugly note at the top that says "You can do this, but you won't be able to save -- we're letting you preview only"
[08:16:28] <Werdna>	or just not ignore any permissions errors
[08:17:22] <MZMcBride>	it's kinda annoying not being able to edit / preview protected pages...
[08:17:34] <Splarka>	heh
[08:17:52] <MZMcBride>	principle of least surprise, i realize
[08:17:54] <TimStarling>	MZMcBride: so you want the current feature, except not broken?
[08:18:01] <Splarka>	could just have a [preview] only button on protected pages (and not write-only)
[08:18:10] <MZMcBride>	a user preference would be nice for that kind of thing
[08:18:15] <MZMcBride>	or what Splarka said
[08:18:26] <Splarka>	can be done via JS I suppose...
[08:18:36] *Werdna 	slaps Splarka 
[08:18:39] *Splarka 	giggles
[08:18:52] <Splarka>	if the devs wern't so slow, userscripts would be mostly obsolete
[08:18:55] *Splarka 	hides
[08:19:02] <Werdna>	I'
[08:19:07] <Werdna>	I've written an essay about that
[08:19:14] *OvrLrd-Q 	hands Splarka a box to hide in
[08:19:16] <MZMcBride>	the Main Page deletion one?
[08:19:31] <Werdna>	bots, user scripts, et cetera, all tend to be responses to a problem that can be solved in MediaWiki.
[08:19:42] <TimStarling>	but it's harder to write software for mediawiki
[08:19:48] <TimStarling>	because mediawiki developers are mean
[08:19:53] <TimStarling>	they review code, and complain when it doesn't work
[08:19:55] <Werdna>	But, most developers are volunteers, and there's insufficient communication in my opinion between developers and users.
[08:20:13] <Werdna>	And yes, there's a lot more quality control and that sort of thing to go through
[08:20:16] <TimStarling>	when you're a user script developer, you can foist your buggy software on anyone, and then you only answer to the affected users
[08:20:19] <Splarka>	and why does the textarea on Special:Undelete sit outside the div? (so it isn't submitted I suppose), and why do those buttons not have IDs (nor the upload button on Special:Upload)
[08:20:28] <MZMcBride>	The biggest issue is that patches sit in Bugzilla for months until they don't work anymore.
[08:20:37] <MZMcBride>	There's no incentive to waste time.
[08:21:28] <TimStarling>	it's a lot easier to ignore a trivial patch which is broken in 5 ways than to explain to the submitter how to fix it
[08:21:39] <Werdna>	right.
[08:21:54] <TimStarling>	if they're interested in further contribution, then it's worth it
[08:22:10] <TimStarling>	but if they're not interested in the project beyond that one patch, it's not
[08:22:19] <MZMcBride>	true
[08:22:54] <MZMcBride>	But sometimes people fire a shot across the bow; no response from one patch will lead people to become discouraged.
[08:23:27] <TimStarling>	well, I did warn everyone on the meta page, didn't I?
[08:23:30] <MZMcBride>	Plus, they see the queue of opened, need-review patches already, and they figure, why bother?
[08:23:33] <TimStarling>	don't submit patches, it's frustrating, I said
[08:23:50] <TimStarling>	ask for commit access instead, that was my advice
[08:24:11] <Werdna>	Yes, that's the good thing about commit access
[08:24:39] <Werdna>	instead of putting a patch there that can be ignored, you shove it in somebody's face so they have three options:
[08:24:42] <TimStarling>				$text .= '<ul class="permissions-errors">' . "\n";
[08:24:44] <Werdna>	(1) Work with you until it's fixed.
[08:24:49] <Werdna>	(2) Hold up scap
[08:24:53] <Werdna>	(3) Revert it.
[08:25:15] <MZMcBride>	Yes, that seems to a be trend recently. Writing decent-ish code and waiting for someone else to fix it.
[08:25:21] <MZMcBride>	Seems to work pretty well, too.
[08:25:42] <TimStarling>	the barrier to entry for commit access is much higher than for submitting patches
[08:25:53] <Werdna>	right
[08:26:08] <MZMcBride>	i thought you had to submit patches in order to get commit access
[08:26:11] <TimStarling>	so we can afford to spend more time helping our committers than we can helping the patchers
[08:26:35] <Werdna>	true
[08:26:37] <TimStarling>	well, generally one patch
[08:26:47] <TimStarling>	then they can commit that themselves
[08:26:56] <MZMcBride>	If the advice truly is to not submit so many patches, that should be in bold on Bugzilla, not "hidden" on a Meta page somewhere.
[08:27:00] *Splarka 	ponders asking for commit access so he can submit patches via wikibits.js, muahaha
[08:27:44] <TimStarling>	if you need to submit more than 2 or 3 patches, you probably need to go away and learn to program for a while before you try again
[08:28:30] <TimStarling>	and of course that's a reason to keep the process somewhat obscure
[08:29:17] <TimStarling>	it means we can save the pride of the crap programmers by pretending that we're rejecting them on the basis of process rather than ability
[08:30:03] <TimStarling>	I'm not talking about 2 or 3 one-line patches though
[08:30:23] <TimStarling>	a patch really has to demonstrate that you are up for the job, that you know how to program, that you're prepared to read the code
[08:30:49] <MZMcBride>	Bugzilla is a black hole; even things marked 'easy' and needs-review are ignored, understandably, but still...
[08:31:01] <TimStarling>	who marks them "easy"?
[08:31:15] <MZMcBride>	Fairies?
[08:31:31] <MZMcBride>	The bug creator or other people, I guess.
[08:32:01] <TimStarling>	I bet it's not the bug creator, I bet it's random people who like to surf bugzilla and organise things
[08:32:22] *Splarka 	has submitted bugs marked 'easy'
[08:32:32] <Splarka>	(but I thought it meant me, not the bug)
[08:32:38] <MZMcBride>	Sometimes it's the bug creator; if it's a capitalization fix or something...
[08:32:39] <TimStarling>	lol
[08:32:43] <Werdna>	Well, it certainly wouldn't be people who knew a lot about php, or they'd be doing the 'easy' patches themselves.
[08:32:59] <TimStarling>	right
[08:33:19] <Splarka>	Werd: how about very easy shell requests? 
[08:33:36] <TimStarling>	shell requests are never easy
[08:33:57] <TimStarling>	in my experience they take 10-15 minutes
[08:34:15] <TimStarling>	very rare to find one that's actually easy
[08:34:36] <TimStarling>	actually making the change of course only takes seconds
[08:34:52] <Werdna>	but I guess you have to check the vote for consensus, or something
[08:35:10] <MZMcBride>	And the obscure spelling of some new namespace or whatever.
[08:35:14] <Splarka>	for, say, changing a rights level, restricting page moves to autoconfirmed for example...what eats the time?
[08:35:17] <TimStarling>	you have to review it for potential impact on the technical side and the community side
[08:35:30] <Splarka>	ahh
[08:35:43] <TimStarling>	Splarka: determining who is asking, why they are asking, how many people are behind them
[08:35:55] <Werdna>	I wonder if we'd do well to have an extension that allows bureaucrats to "approve" a shell request, and then a sysadmin can implement or refuse to implement it.
[08:36:23] <MZMcBride>	An extension? Sounds like a noticeboard would suffice.
[08:36:31] <flyingparchment>	Werdna: that can be done from any decent bug tracker, to
[08:36:33] <flyingparchment>	too
[08:36:35] <Werdna>	Or just, as a matter of policy, require the say-so of a bureaucrat before going through with one.
[08:36:59] <Werdna>	that way we don't have things like poor jens copping flak for the rollback thing
[08:36:59] <AaronSchulz>	hmm
[08:37:03] <MZMcBride>	Well, during the last en.wiki fiasco, I think Jimbo suggested ArbCom.
[08:37:13] <AaronSchulz>	human shields eh?
[08:37:35] <Werdna>	and sysadmins don't need to keep up with community practice to figure out whether they're going to be shot for approving one.
[08:38:04] <MZMcBride>	Well, I imagine 95% of shell requests are entirely uncontroversial.
[08:38:06] <flyingparchment>	oddly enough, the easiest shell requests are probably new wiki creations
[08:38:13] <MZMcBride>	En.wiki will always find a reason to bitch.
[08:38:14] <flyingparchment>	since that's the one thing that has a well defined procedure
[08:38:23] <Werdna>	It transfers the responsibility of determining a consensus (and thus any blame for it) to people other than those who happen to have the technical ability to go through with them.
[08:39:15] <TimStarling>	MZMcBride: how do you tell them from the other 5%?
[08:39:19] <Splarka>	https://bugzilla.wikimedia.org/show_bug.cgi?id=10316  <-- bah, took me a while to find it, I marked it as easy because it involved a tiny change to the gen=js (though the actual fix was more complex)
[08:39:45] <MZMcBride>	TimStarling: make sure the site requesting isn't the English Wikipedia
[08:39:53] <Werdna>	touche.
[08:40:03] <TimStarling>	ah yes, because the other wikis are utopias where everyone gets along
[08:40:36] <TimStarling>	if there's only one person active on a wiki, then it's pretty easy
[08:40:47] <TimStarling>	but then there's no review
[08:40:48] <AaronSchulz>	the Army of One
[08:40:49] *Splarka 	ponders saying "subst/gets along/speaks in non-english so their bitchings sound like white noise/" but thinks it would be controvertial
[08:40:57] <TimStarling>	what if that one person is on crack?
[08:41:02] <AaronSchulz>	the One Man Band
[08:41:07] <TimStarling>	so then it falls to you to review what they are asking for
[08:41:39] <wikibugs>	03(NEW) Add import source to de.wiki - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13089 15enhancement; normal; Wikimedia: Site requests; (raimond.spekking)
[08:41:46] <wikibugs>	03(NEW) Add magic word to generate aggregate TOC from a set of pages - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13090 15enhancement; normal; MediaWiki: Page rendering; (spidermannequin)
[08:41:55] <MZMcBride>	Yes, sysadmins are always concerned with the crack-using obscure-language-editing user who files Bugzilla requests.
[08:41:55] <_wooz>	lo
[08:42:17] <TimStarling>	"You do not have permission to do that, for the following reason:
[08:42:17] <TimStarling>	You are not allowed to execute the action you have requested.
[08:42:17] <TimStarling>	"
[08:42:22] <TimStarling>	...informative
[08:42:44] <Splarka>	at least the locked-db error message is humorously informative ($1)
[08:42:52] <MZMcBride>	heh
[08:43:01] <Splarka>	or, uninformative, rather
[08:43:38] <Werdna>	TimStarling: That would appear if the action requires a permission key which isn't assigned to any groups, I think.
[08:44:23] <AaronSchulz>	Splarka: give us a $1 and we will tell you why the db is locked
[08:44:29] <TimStarling>	it's referenced from 5 places in Title.php
[08:44:58] <MZMcBride>	heh. a PHP joke
[08:46:23] *Splarka 	donates one yen
[08:47:05] <AaronSchulz>	reminds me
[08:48:19] *AaronSchulz 	donates the value of the copper in a penny to Splarka
[08:48:41] <AaronSchulz>	heh, how much is a penny really worth?
[08:48:55] <AaronSchulz>	mmm...equivocation
[08:48:55] <MZMcBride>	well, it costs almost 2 cents to make one
[08:49:02] <AaronSchulz>	;)
[08:49:05] <TimStarling>	alright, time to commit this thing
[08:49:07] <MZMcBride>	and it's illegal to melt them down anymore
[08:49:15] <Werdna>	lols
[08:49:19] <TimStarling>	I'm using editredlink, a compromise between me and brion
[08:50:17] <TimStarling>	I did some quick testing of user permissions errors in EditPage.php by using a stub getUserPermissionsErrors()
[08:50:32] <Werdna>	TimStarling: should redlinks have a tooltip to the effect of "Click here to make this article"?
[08:50:43] <TimStarling>	just a little function that returns anything you want, then you can edit it to see what different return values do
[08:51:08] <TimStarling>	Werdna: they could quite easily have such a tooltip
[08:51:22] <TimStarling>	but that would conflict a bit with the current feature...
[08:51:38] <TimStarling>	and the user might not see the tooltip anyway, so it wouldn't replace the current feature
[08:51:47] <Werdna>	right
[08:52:11] <Werdna>	I wasn't thinking of it replacing the feature, simply that it might save the odd click and back-button on the browser.
[08:53:01] <TimStarling>	maybe "nonexistent article" then?
[08:53:11] <TimStarling>	what it is, rather than what the red link may or may not do
[08:53:27] <Werdna>	in brackets after the title, perhaps?
[08:53:57] <Werdna>	[[asdfhkjiyhiuof]] produces a tooltip of "asdfhkjiyhiuof (not yet written)"
[08:53:59] <Werdna>	or something
[08:54:57] <TimStarling>	that sounds nice
[08:55:12] <TimStarling>	I'll add that
[08:55:57] <TimStarling>	'and'            => 'and',
[08:56:02] <Werdna>	but "not yet written" is probably poorly worded
[08:56:06] *TimStarling 	covers face
[08:56:13] <TimStarling>	talk about cultural assumptions
[08:56:29] <TimStarling>	I hope that's not actually used anywhere
[08:58:23] <TimStarling>	I like "not yet written" better than "nonexistent article"
[08:58:37] <TimStarling>	too many syllables in "nonexistent"
[08:59:52] <Nikerabbit>	4? :)
[09:00:14] <TimStarling>	yes, quite a mouthful
[09:00:36] <Nikerabbit>	sounds like a normal length of finnish words :)
[09:03:11] *Splarka 	ponders mentioning...
[09:03:37] <Splarka>	editredlink assumes nonexistant links are in fact red (they could be styled differently, or use that silly questionmark pref)
[09:04:28] <TimStarling>	yes, and it assumes that the person reading it speaks english
[09:04:38] <Splarka>	touche` as well
[09:04:53] <puzzlet_>	how about "red"? ;)
[09:05:14] <TimStarling>	they weren't called red links originally, they were called broken links
[09:05:24] <TimStarling>	perhaps because they weren't always red in the original software
[09:05:28] <CIA-6>	03siebrand * r31137 10/trunk/extensions/GoogleMaps/ (11 files): (partial) re-indent and EOL whitespace removed
[09:05:47] <TimStarling>	but now we think of them as red links because they are red in real life, and it helps to draw your naming conventions from real life
[09:05:57] <TimStarling>	easier to understand and remember
[09:06:23] <TimStarling>	in usemod they were blue question marks after the link text
[09:06:37] <Splarka>	US "pennies" ("cent" is the only official name) are worth 1/100th of a dollar as currency, about 1.2 cents as scrap metal (copper plated zinc currently, pre 1982 pennies were made of brass and are worth about 2 cents), and cost as stated about 2 cents to make... but the mint only produces money on demand, so they aren't really losing money, but could be if people demanded them just to melt them down (hence the recent ban
[09:06:49] <Splarka>	... on melting of cent and nickel coins in the US, as well as exporting limits
[09:07:15] <puzzlet_>	and "broken" doesn't seem to be appropriate to me as well, since creating the page that link points to doesn't really feel like "fixing" the link
[09:07:36] <Splarka>	they should either reformulate them or ditch them completely, but 2009 is the upcoming 100th anniversary of the lincoln cent, so they have plans for some design changes
[09:08:40] <puzzlet_>	we did that in South Korea last year - a new 10-Won coin in aluminum
[09:08:47] <TimStarling>	what do you think about https://bugzilla.wikimedia.org/show_bug.cgi?id=542 ?
[09:08:58] <TimStarling>	should I fix that too, while we're on the subject of links?
[09:09:01] <Splarka>	so tim, which shade of red is official? 
[09:09:09] <Splarka>	/monobook/main.css says:
[09:09:13] <Splarka>	a.new, #p-personal a.new {
[09:09:13] <Splarka>		color: #ba0000;
[09:09:25] <Splarka>	but gen=css (if you have selected it) declares:
[09:09:29] <Splarka>	a.new, #quickbar a.new { color: #CC2200; }
[09:09:46] <TimStarling>	*shrug*
[09:09:50] <Splarka>	this was only really an issue on memory alpha that I can remember, on a black background it is easy to spot the difference
[09:09:55] <TimStarling>	is this more important than bug 542?
[09:10:23] <TimStarling>	a long and complex bug...
[09:10:30] <TimStarling>	controversial
[09:10:41] <wikibugs>	03(mod) Quiz extension text responses not markable in MW1.12a - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13074  (10mccormack)
[09:10:50] <TimStarling>	but never closed
[09:10:52] <Splarka>	just sayin, maybe they should be called orangelinks
[09:11:01] <TimStarling>	you'd think brion would have closed it if he disagreedw ith it
[09:11:11] <Splarka>	because pedantic semantics are the makeup of fanatics
[09:13:04] <flyingparchment>	TimStarling: _if_ there was some sort of indication that a title attribute existed, without having to hover over the link, i'd accept that having them in that case is redundant.  otherwise, i don't see the problem
[09:14:12] <TimStarling>	flyingparchment: the theory is that maybe there's a client with exactly those properties
[09:14:41] <TimStarling>	the screen reader JAWS apparently reads out title attributes, except (maybe as a special-case hack just for us) if the title attribute is the same as the link text
[09:14:59] <TimStarling>	so maybe some less enlightened screen readers lack that hack
[09:15:14] <CIA-6>	03siebrand * r31138 10/trunk/extensions/GoogleMaps/GoogleMaps.i18n.php: 
[09:15:14] <CIA-6>	* update i18n formatting
[09:15:14] <CIA-6>	* add from http://www.mediawiki.org/wiki/Extension:Google_Maps/Languages/BR
[09:15:14] <CIA-6>	* add from http://www.mediawiki.org/wiki/Extension:Google_Maps/Languages/AR
[09:18:22] <CIA-6>	03siebrand * r31139 10/trunk/extensions/GoogleMaps/ (GoogleMaps.body.php GoogleMaps.i18n.php): Fix a few syntax errors
[09:27:18] <TimStarling>	ok, title attribute staying as is for now, but I'm putting the informative text in the title
[09:27:47] <TimStarling>	sorry I mean bug 542 not solved
[09:33:11] <CIA-6>	03tstarling * r31140 10/trunk/phase3/ (7 files in 3 dirs): 
[09:33:11] <CIA-6>	* Don't show edit permissions errors on a red link click, just redirect to the article. This is so that readers who don't know what a red link is aren't confused when they are told they are range-blocked.
[09:33:11] <CIA-6>	* Put "not yet written" in the title attribute of red links, so that readers unfamiliar with the site might guess what the colour means.
[09:33:11] <CIA-6>	* Fixed minor bugs and refactored code in the vicinity of EditPage::edit().
[09:34:27] <TimStarling>	ok, what's next?
[09:36:30] <CIA-6>	03siebrand * r31141 10/trunk/extensions/GoogleMaps/GoogleMaps.php: Some more formatting
[09:39:04] <wikibugs>	03(FIXED) "Permission error" nasty language for innocent red link clicker - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13058  +comment (10tstarling)
[09:42:50] <TimStarling>	12206?
[09:42:53] <emilsedgh>	i need some help for customizing mediawiki.first, i want to make [[Image]] tag, goes to 'thumb' mode automatically.i looked through the code, find out that the Thumbnails are shown using Linker class, but i didnt found out where the [[Image]] is being parsed
[09:45:05] <TimStarling>	emilsedgh: in Linker::makeImageLink2(), there's a parameter called $handlerParams
[09:45:28] <CIA-6>	03siebrand * r31142 10/trunk/extensions/ (3 files in 2 dirs): Add support for GoogleMaps extension to Translate
[09:46:10] <TimStarling>	at line 595, there is this:
[09:46:14] <TimStarling>			if ( $file && $hp['width'] ) {
[09:46:15] <TimStarling>				# Create a resized image, without the additional thumbnail features
[09:46:15] <TimStarling>				$thumb = $file->transform( $hp );
[09:46:34] <TimStarling>	$hp is short for $handlerParams
[09:46:42] <TimStarling>	basically before that point, you can set the width parameter
[09:46:48] <emilsedgh>	hm
[09:47:01] <TimStarling>	if ( !isset( $hp['width'] ) ) $hp['width'] = 200;
[09:47:12] <TimStarling>	then if there's no width set, the width will be 200
[09:47:39] <TimStarling>	but don't set it before the optional call to makeThumbLink2(), because makeThumbLink2() has its own size defaults if none are specified
[09:47:43] <TimStarling>	and you don't want to override it
[09:48:44] <emilsedgh>	TimStarling: one little thing, i even add die() to these functions, but its not affecting.is there any kind of caching or something like that?
[09:48:52] <Splarka>	Tim: how about http://en.wikipedia.org/w/index.php?title=Wikipedia:Village_pump_%28technical%29#Error_message (parser bug?) if your goal is to disconfuszilate people
[09:48:56] <TimStarling>	yes
[09:49:11] <TimStarling>	better to test it using the edit preview
[09:49:36] <TimStarling>	saving the page will clear the cache, even if you don't make any changes
[09:49:45] <emilsedgh>	yah
[09:49:48] <emilsedgh>	thanks
[09:52:07] <CIA-6>	03siebrand * r31143 10/trunk/extensions/Translate/Translate.php: Comment out GoogleMaps. Have to look into this...
[09:56:59] <emilsedgh>	TimStarling: is there any way to do such customizations without touching core files of mediawiki?
[10:01:11] <TimStarling>	not that particular customisation, no
[10:01:26] <TimStarling>	the convention is to use hooks, but I don't think there are any relevant hooks for that
[10:02:30] <TimStarling>	so the most robust thing you could do is add a hook, submit a patch for the addition of the hook, and make your customisation use the hook
[10:02:55] <TimStarling>	hmm
[10:03:02] <TimStarling>	or you could subclass the skin
[10:03:03] <emilsedgh>	hm
[10:03:31] <TimStarling>	you could make your own skin with its own version of makeImageLink2
[10:03:46] <Splarka>	or you could fake it... what aspect of |thumb do you want? the border? the little zoom thingy?
[10:03:56] <TimStarling>	that wouldn't require core patches, but maybe a bit of repeated code
[10:04:02] <emilsedgh>	TimStarling: another question, sorry for bugging, but it looks like that when 'thumb' is not set, this function is not being called....who calls it?
[10:04:36] <emilsedgh>	both, Splarka, i also want to make it go left as default (this one was easy i did it ::P)
[10:04:44] <TimStarling>	Splarka: the bug isn't a parser bug, it's a bug with the permissions code in Article.php
[10:05:14] <TimStarling>	Parser::makeImage() calls it
[10:05:15] <Splarka>	Tim: ahh well, still funny (per VOA...pay us to find out)
[10:05:41] <Splarka>	emil: you could sort of do both in CSS (except the sizing)
[10:06:58] <emilsedgh>	Splarka: i dont mind doing the hard thing, i want the right thing
[10:06:59] <emilsedgh>	oh
[10:07:00] <CIA-6>	03raymond * r31144 10/trunk/phase3/maintenance/language/messages.inc: Update per r31140.
[10:07:24] <emilsedgh>	there is a hook: 'BeforeParserMakeImageLinkObj', maybe this could help me?
[10:08:20] <TimStarling>	who wrote this code anyway?
[10:08:34] <Splarka>	well, one thing, what if they omit a caption? [[Image:Foo.jpg|thumb]] looks pretty ugly
[10:09:08] <Splarka>	(the little zoom icon gets squished by the bottom border in some (all?) browsers)
[10:09:23] <TimStarling>	was it more aaron/simetrical teamwork?
[10:09:40] <TimStarling>			if( in_array( array( 'blocked' ), $result ) ) {
[10:09:44] <emilsedgh>	Splarka: i want [[Image:Foo.jpg]] adds thumb|left autimatically.i also want the image summary goes there as default caption (if there is no other caption set)
[10:10:08] <Splarka>	ahh, fair enough
[10:10:38] <TimStarling>	ok, question, do we really really need to be able to have multiple errors with the same message name?
[10:11:33] <TimStarling>	emilsedgh: BeforeParserMakeImageLinkObj is too early, and it doesn't let you change $parts or $options
[10:11:41] <emilsedgh>	yes
[10:12:33] <emilsedgh>	TimStarling: what i have to do to add a hook? adding wfRunHooks('FooBar', $p1, $p2) is enough? 
[10:12:40] <Splarka>	creeping of "Hmm, I need the word 'edit' for this patch, and translated in as many languages as possible, oh I know, wfMsg('edit')! solved" ?
[10:13:06] <TimStarling>	wRunHooks('FooBar', array($p1, $p2))
[10:13:19] <emilsedgh>	good
[10:13:25] <emilsedgh>	thanks TimStarling :)
[10:13:37] <TimStarling>	well, &$p1 if you want to change it
[10:13:49] <TimStarling>	references in arrays need to be explicit at the caller
[10:14:31] <Nikerabbit>	Splarka: be carefull with reuse
[10:15:28] <Splarka>	Niker: it was a caricaturesque quote of a lazy dev
[10:15:29] <emilsedgh>	TimStarling: so, adding wfRunHooks( 'BeforeParserCallsImageLinkObj', array( &$params ) ); in line 4626 could help people like me i think, and probably it will be good if you commit it there.
[10:19:18] <TimStarling>	how about...
[10:19:22] <TimStarling>	wfRunHooks( 'ParserMakeImageParams', array( $title, $file, &$params ) );
[10:20:20] <emilsedgh>	im thinking about moving it a few lines above, because in that line, the default values are set, so the hook doesnt knows about parameters sent by user
[10:20:52] <emilsedgh>	oh
[10:20:53] <emilsedgh>	no
[10:21:04] <emilsedgh>	TimStarling: yeah, thats good, it will be so nice if you commit that
[10:21:21] <emilsedgh>	TimStarling: so i wouldnt need to patch again the mediawiki in next update ;)
[10:21:31] <TimStarling>	well, your lines are different to my lines..
[10:21:47] <TimStarling>	I've got this:
[10:21:48] <TimStarling>			$params['frame']['alt'] = $alt;
[10:21:48] <TimStarling>			$params['frame']['caption'] = $caption;
[10:21:48] <TimStarling>			wfRunHooks( 'ParserMakeImageParams', array( $title, $file, &$params ) );
[10:21:48] <TimStarling>			# Linker does the rest
[10:21:50] <TimStarling>			$ret = $sk->makeImageLink2( $title, $file, $params['frame'], $params['handler'] );
[10:22:50] <emilsedgh>	TimStarling: thats good
[10:23:01] <emilsedgh>	so im using yours
[10:23:04] <TimStarling>	ok, I'll commit it later
[10:24:21] <emilsedgh>	TimStarling: so, could i make sure that this will be commited? if no i have to store a patch somewhere so i dont forget to add it on next update
[10:24:46] <TimStarling>	if you insist
[10:25:31] *emilsedgh 	insists ;)
[10:25:45] <emilsedgh>	TimStarling: btw, thanks for help, i really appreciate it
[10:25:56] <CIA-6>	03tstarling * r31145 10/trunk/phase3/includes/Parser.php: Hook suggested on IRC.
[10:26:12] <TimStarling>	is that sure enough?
[10:26:42] <emilsedgh>	i think yes
[10:27:00] <emilsedgh>	;)
[10:35:04] <emilsedgh>	how could i get the 'summary' of the files? the $file sent to that hook hasnt that...
[10:37:21] <wikibugs>	03(mod) interwiki [[tlh:]] does not work correctly - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=9164  +comment (10malafaya)
[10:45:27] <CIA-6>	03siebrand * r31146 10/trunk/extensions/GoogleMaps/ (EditorsMap.js GoogleMaps.i18n.php): Rename message 'note' to 'gm-note'
[10:57:50] <Werdna>	TimStarling: didn't you once tell me about a time you hacked a floppy disk drive to act as a way to transfer files between computers? As in a 5 3/4 inch floppy drive.
[11:01:14] <flyingparchment>	Werdna: i believe you can hack a floppy drive that way by putting a disk in, copying the files to it and moving the disk to the other computer
[11:01:52] <Werdna>	Yes, I think tim told me something about connecting the read wire from one to the write wire from the other, or something like that, many years ago
[11:01:56] <Werdna>	without a disk
[11:01:59] *Werdna 	shrugs
[11:02:13] <Werdna>	just a curiosity.
[11:02:28] <flyingparchment>	i have an old ftape drive around here somewhere, that was pretty neat
[11:02:38] <flyingparchment>	(a tape drive that had a standard 3.5" floppy connector, so you could use it anywhere)
[11:03:06] <Werdna>	I felt so old today when I saw a younger kid's homework question: "What was a floppy disk?"
[11:03:17] <Splarka>	http://xkcd.com/378/
[11:05:45] <TimStarling>	Werdna: no, a serial port
[11:06:30] <TimStarling>	a serial port, a bit of speaker wire for TX/RX, and a bit of stranded wire for ground
[11:06:44] <Werdna>	and connected to what on the other end?
[11:06:49] <steffkes>	hey .. short question .. thumbnails for pdf? already using imagemagick for the image-thumbnails
[11:06:56] <TimStarling>	ah yes, this is where the disks come in
[11:07:18] <TimStarling>	I had an XT with 5.25" disk, and an atari ST with a 3.5" disk
[11:07:32] <TimStarling>	I was copying from one computer to the other, because I needed some data off the XT
[11:07:36] <Splarka>	so you made a null modem?
[11:07:40] <TimStarling>	yes
[11:07:51] <Werdna>	clever.
[11:08:03] <TimStarling>	with my own file transfer program written in the BASIC dialect of each computer
[11:08:21] <Splarka>	there was a shareware program on the BBSes to do that for you
[11:08:40] <Werdna>	so you connected them serial to serial?
[11:08:42] <Splarka>	it would copy itself (if you could get it to one machine) if you typed a pipe command on the other
[11:08:51] <flyingparchment>	i miss computers nowadays not having serial ports, they're so easy to connect for when you need to move data right now, and don't have any other cables
[11:09:09] <flyingparchment>	(you don't even need a soldering iron if you just twist the cable around the connector on the plug)
[11:09:28] <TimStarling>	Werdna: yes
[11:09:43] <Werdna>	ah
[11:09:54] *Werdna 	notices that his computer has a serial port that he never noticed before.
[11:10:08] <TimStarling>	flyingparchment: in this case I had some little plastic connector sleeves meant for connecting car auto wires, I think
[11:10:16] <TimStarling>	they fit neatly around the serial pins
[11:10:18] <Splarka>	had brother's laptop (386 or 486?) with a serial port, and had my desktop with a serial port and a modem (2400), wrote a BASIC program to reroute the traffic from serial-serial (null) to modem, the laptop thought it had a modem, could even play DOOM over it, and spy on the packets
[11:10:24] <TimStarling>	just wrapping the wire around was kind of unreliable
[11:10:45] <Werdna>	I find the lo-fi stuff that lets you tinker that remains on unix-like systems.
[11:11:08] <domas>	yawn
[11:11:11] <Splarka>	(since he had a much faster machine but no modem, and doom was very slow on the desktop)
[11:11:21] <Werdna>	If you know your way around it (I don't claim to be excellent at this) then all this stuff is really versatile.
[11:11:31] <TimStarling>	car electrical wires, I should say
[11:11:36] <Werdna>	and really quick to work with
[11:13:11] <TimStarling>	the C64 (I never owned one) had a user port, with a whole lot of logic outputs that you could switch on and off in software
[11:13:30] <TimStarling>	it was great for gaming, but it obviously had some hobbyist features as well
[11:14:25] <TimStarling>	these days, if all you had was USB, presumably you'd have to shell out for some kind of trainer board or something
[11:15:00] <TimStarling>	they still have a joystick port don't they?
[11:15:02] <flyingparchment>	i believe you can get PCI cards that allow some level of direct bus access
[11:15:26] <flyingparchment>	joystick ports are going away now too, with usb joysticks
[11:15:26] <TimStarling>	joystick ports are very cool things for hobbyists
[11:15:34] <TimStarling>	pity
[11:16:13] <TimStarling>	one of my friends in high school made a weather station for a science project, using the joystick port ADC
[11:16:42] <TimStarling>	he may have even used a disassembled mouse for windspeed, I'm not sure...
[11:17:13] <Werdna>	I wish I could do this sort of stuff now - I've been thinking about turning my old laptop into a new alarm clock with configurable days and such - but I need some way of being able to press a button to turn it off without taking out the laptop and opening it up
[11:17:47] <flyingparchment>	Werdna: use a usb mouse?  the circuitry inside those is very small
[11:17:56] <flyingparchment>	and you get 5 buttons
[11:18:06] <TimStarling>	yeah, you can still disassemble mice at least
[11:18:15] <TimStarling>	but you'd have trouble measuring windspeed with an optical mouse
[11:18:32] <flyingparchment>	heh
[11:18:51] <flyingparchment>	maybe not, use a weather cock and have it slide something over the sensor
[11:19:06] <Werdna>	flyingparchment: I'm not sure how to accept button inputs into my own program, instead of having it just make the mouse on the computer click - although, I could cover the screen with a window and accept the clicks through software
[11:19:20] <Splarka>	they make usb "buttons"
[11:19:23] *Werdna 	is, however, a newbie in the unix sense (has not written more than one device driver)
[11:19:28] <flyingparchment>	Werdna: under Unix, you can read the raw data from the mouse using a /dev device
[11:19:29] <Splarka>	(if the laptop has usb)
[11:19:44] <Werdna>	it does have usb, and it's running kubuntu at the moment.
[11:19:47] <Werdna>	so maybe I'll look into it
[11:19:50] <flyingparchment>	Werdna: don't even need a device driver, it's in userland
[11:19:56] <Werdna>	yay
[11:20:09] <Werdna>	I'll have to find a way to disable it being used as an actual mouse
[11:20:15] <Splarka>	http://www.google.com/products?um=1&hl=en&c2coff=1&safe=off&q=USB+button&ie=UTF-8&sa=N&tab=if
[11:20:15] <Werdna>	but I'm sure I can figure it out
[11:20:18] <flyingparchment>	just don't start X or gpm
[11:20:32] <Duesentrieb>	/dev and /proc are among the things i *really* like about *nix systems
[11:20:35] <Werdna>	X is usually running, though.
[11:20:42] <flyingparchment>	Duesentrieb: linux /proc is awful though
[11:20:47] <TimStarling>	you want to run X on an alarm clock?
[11:20:51] <flyingparchment>	(with all the overloaded stuff they added)
[11:20:59] <AmeedX>	Hi Friends .. Can i help
[11:21:08] <Duesentrieb>	flyingparchment: well, as always: the idea is good. what they make of it is faky
[11:21:22] <Werdna>	I'm already using the laptop for something. I figure if I just remember to plug the USB mouse in when I'm finished with it, it'll work as an alarm clock
[11:22:36] <TimStarling>	otherwise you'll just have to find the mouse of a morning to turn it off
[11:22:58] <Werdna>	that's not necessarily a bad thing. It'd force me to get out of bed :P
[11:23:03] <TimStarling>	anyway, you wouldn't use an actual mouse
[11:23:22] <Werdna>	what would I use?
[11:23:29] <Splarka>	http://www.pocket-lint.co.uk/news/news.phtml/7990/9014/USB-Smart-Button-USB-Gadget.phtml
[11:23:41] <TimStarling>	have you got a soldering iron?
[11:23:48] <Werdna>	Yes.
[11:23:58] <TimStarling>	I mean, you could use a mouse if you liked the look of it beside your bed...
[11:24:03] <Werdna>	I've even used it occasionally!
[11:24:05] <TimStarling>	but otherwise, you could easily make your own switch
[11:24:22] <TimStarling>	you'd take a mouse, take it apart, maybe replace the USB cable with a longer one if need be
[11:24:48] <TimStarling>	cut the existing switches off, and solder your own wires on in their place
[11:25:10] <TimStarling>	then you can connect the other end to your own momentary switch
[11:25:18] <TimStarling>	or whatever switch you want
[11:25:26] <Werdna>	just a little push-button thing, a few dollars from jaycar
[11:25:54] <TimStarling>	yep, whatever meets your tastes
[11:26:48] <Werdna>	hmmm.. I'm going to end up doing this now.
[11:26:50] <Duesentrieb>	hm... those smart usb buttons...
[11:26:52] <Splarka>	or how about that 3 button Optimus keyboard
[11:26:57] <Duesentrieb>	that would be good wikipedia merchandize
[11:26:57] <Werdna>	heh
[11:26:58] <Splarka>	only $162
[11:27:05] <Duesentrieb>	you know, and EDIT button. a FAT one.
[11:27:14] <Werdna>	nah, random article
[11:27:14] <TimStarling>	you could keep the existing switches I guess, with the circuit board in another case, it's just that they're finicky little things
[11:27:18] <TimStarling>	they might be hard to work with
[11:27:28] <Werdna>	sell it with software to open a new random article when you push it
[11:27:41] <Werdna>	Is it possible to stop X from seeing it as a mouse, though?
[11:28:07] <TimStarling>	I used to have a lot of trouble getting X to *start* seeing my mouse as a mouse
[11:28:09] <flyingparchment>	Werdna: linux provides a separate /dev entry for each mouse, and one that combines events from all mice.  tell X to use the specific one for your other mouse/trackpad instead of the combined one
[11:28:14] <TimStarling>	surely it couldn't be that hard to make it stop
[11:28:20] <Duesentrieb>	Werdna: i suppose you can tell it in xorg.conf what do use as a mouse
[11:28:22] <Werdna>	:)
[11:28:29] <Duesentrieb>	like, set it to use the touchpad only or something
[11:28:42] <Splarka>	http://sewelldirect.com/Optimus-Mini-Three-V-20-3-Key-Programmable-Keyboard-Aluminum.asp
[11:29:10] <flyingparchment>	Splarka: $162?
[11:29:36] <flyingparchment>	that's an expensive alarm clock
[11:29:40] <Splarka>	fp: http://en.wikipedia.org/wiki/Optimus_Maximus_keyboard
[11:29:45] <Werdna>	        Option          "Device"                "/dev/input/mice"
[11:29:48] <Splarka>	some famous vaporware
[11:29:48] <Werdna>	excellent
[11:30:05] <Splarka>	damn the expense! heh
[11:30:23] <Splarka>	how much is a mouse + laptop compared to a $2 LCD alarm clock?
[11:30:46] <flyingparchment>	well, i already have an old broken mouse and a laptop, so $0
[11:31:14] <Werdna>	yeah, but does a $2 LCD alarm clock allow configuration for every day of the week, potentially synchronising with calendars, reading out information in the morning once it wakes you up, etc, etc?
[11:31:41] <Splarka>	yep, it uses the "set it when you go to bed" interface
[11:31:50] <Splarka>	and "leave yourself a note"
[11:31:53] <Duesentrieb>	Werdna: that's called a "cell phone".
[11:32:07] <flyingparchment>	my phone can't do that :(
[11:32:18] <Werdna>	my phone probably could, but it would be clunky
[11:32:32] <Werdna>	and it wouldn't be DIY!
[11:32:55] <TimStarling>	Werdna, did you invent getUserPermissionsErrors()?
[11:33:39] <Splarka>	other solutions.. USB webcam with motion sensor software that shuts off when you stand up... USB infrared thermometer that shuts off when there is no heat source in your bed...
[11:33:43] <Werdna>	TimStarling: yes.
[11:34:13] <TimStarling>	the data structure that it returns is kind of awkward, a few people are having trouble using it
[11:34:31] <Werdna>	how should it return the data?
[11:34:34] <TimStarling>	do you know of any reason why you'd want to have multiple permission errors with the same message name?
[11:34:47] <Werdna>	not really
[11:34:54] <flyingparchment>	TimStarling: i found a bug in WebRequest::getRequestURL(), let me tell you about it
[11:35:00] <TimStarling>	because one fairly simple format would be to have the message name in the key
[11:35:03] <Werdna>	so I suppose you could do array( 'messagename' => array( parameters) );
[11:35:08] <TimStarling>	yes, that
[11:35:39] <Werdna>	The intention is for it to be done with call_user_func_array( 'wfMsg', $permErrors );
[11:35:45] <TimStarling>	otherwise we'd have to have a special class to handle the various operations people want to do on that array
[11:35:47] <flyingparchment>	TimStarling: if REQUEST_URI is not set, it uses SCRIPT_NAME?QUERY_STRING; but if you're using action paths, /history/Whatever becomes /history/Whatever?title=whatever&action=history&otherparams
[11:35:52] <Werdna>	well, that was my thinking when I did it like last June
[11:36:33] <TimStarling>	flyingparchment: I'll have a look at it
[11:36:58] <flyingparchment>	you can see the problem here: http://wiki.ts.wikimedia.org/view/Special:Version on the "special" link
[11:37:08] <TimStarling>	Werdna: the class would be simpler if it could have unique message names
[11:37:29] <TimStarling>	the main thing people want to do is check whether a particular message name is present, and remove particular message names
[11:37:40] <Werdna>	true.
[11:37:41] <TimStarling>	so it makes sense to have it indexed by message name
[11:37:47] <Werdna>	right.
[11:38:18] <Werdna>	well, to ensure backwards-compatibility, [messagename] => array( [messagename], parameters ... ) makes sense.
[11:38:40] <TimStarling>	maybe if the index is numeric, it could switch to the old format
[11:38:54] <TimStarling>	that's ugly though...
[11:39:09] <TimStarling>	is there a need for backwards compatibility?
[11:39:31] <Werdna>	Depends. How many extensions rely on the behaviour (it's 9-month-old code)
[11:40:07] <TimStarling>	maybe grep can tell us
[11:40:38] *Werdna 	has a look
[11:40:45] <TimStarling>	are there hooks?
[11:40:56] <TimStarling>	mmm, yes, there are hooks
[11:41:04] <TimStarling>	that kind of implies the existence of extensions doesn't it?
[11:41:14] <Werdna>	well, I put the hooks in as a matter of course
[11:41:27] <Werdna>	I figured it was almost certainly something that would be hooked.
[11:47:22] <TimStarling>	freaky
[11:47:32] <TimStarling>	there's an extension called RightFunctions
[11:47:55] <Werdna>	whee
[11:48:01] *Werdna 	has the PCB of an optical mouse on his desk
[11:49:24] <CIA-6>	03siebrand * r31147 10/trunk/extensions/GoogleMaps/ (5 files): 
[11:49:24] <CIA-6>	Rename a lot of messages.
[11:49:24] <CIA-6>	* could not find out exact usage of some, which were not renamed (map, width, height, meters, miles)
[11:49:24] <CIA-6>	* untested. In contact with author for further updated.
[11:50:44] <TimStarling>	TitleBlacklist is the only one that would be broken by using an associative array structure
[11:52:04] <TimStarling>	ImportFreeImages, Metavid and RightFunctions call getUserPermissionsErrors(), they assume it's an array, but don't interpret the members at all
[11:52:05] <Werdna>	because it would say "You've violated blacklist items X,Y,Z"?
[11:52:27] <TimStarling>	yes
[11:52:28] <Emily2>	Hi!
[11:52:31] <TimStarling>					$result = array( $message,
[11:52:31] <TimStarling>						htmlspecialchars( $blacklisted->getRaw() ),
[11:52:31] <TimStarling>						$title->getFullText() );
[11:52:46] <Werdna>	but it's a trivial fix
[11:52:52] <Emily2>	maybe TimStarling can help me :)
[11:52:58] <Werdna>	maybe he can
[11:53:04] <flyingparchment>	Werdna: as long as the fix doesn't break it for old versions
[11:53:04] <Werdna>	depends on what you need help with
[11:53:10] <Emily2>	I want to create a printable HTML version of a complete book and I don't know how
[11:53:11] <Werdna>	flyingparchment: ah, right.
[11:53:22] <Werdna>	Emily2: and you want to use MediaWiki?
[11:53:27] <Emily2>	http://ro.wikibooks.org/wiki/Wikijunior:Alfabetul_animal
[11:53:40] <Werdna>	ah, a wikibook
[11:53:44] <Emily2>	no one wants to help from #wikibooks
[11:53:56] <TimStarling>	that sounds really really complicated
[11:54:01] <Werdna>	well, it is possible to download an HTML version of each individual page.
[11:54:11] <Emily2>	I don't think so :)
[11:54:25] <Werdna>	sure
[11:54:49] <Emily2>	en.wikibooks has a printable version
[11:55:06] <Emily2>	I only want one for ro.wikibooks
[11:55:10] <Werdna>	Emily2: try this: http://ro.wikibooks.org/wiki/Wikijunior:Alfabetul_animal?useskin=none
[11:55:36] <Duesentrieb>	Emily2: there's a full featured service for this upcomming, but it's not ready yet. see http://wikimediafoundation.org/wiki/Wikis_Go_Printable
[11:55:42] <Werdna>	oops
[11:55:50] <Werdna>	wrong link
[11:56:03] <Werdna>	http://ro.wikibooks.org/wiki/Wikijunior:Alfabetul_animal?action=render
[11:56:06] <Werdna>	that's the one
[11:56:22] <Duesentrieb>	Werdna: that's one page. not the full book.
[11:56:43] <Werdna>	yeah, so do it for each page
[11:57:07] <Duesentrieb>	then assemble, convert to pdf, fix pagination, set it, and the print it.
[11:57:17] <Duesentrieb>	and then curse because the thumbnail resolution sucks
[11:57:18] <Emily2>	sounds complicated
[11:57:22] <Duesentrieb>	it is
[11:57:26] <Emily2>	but I am going to try
[11:58:09] <Werdna>	hehe
[11:58:13] <Duesentrieb>	Emily2: PediaPrint sounds good, i hope it will *be* good too. there were some wiki-to-pdf services before, nto sure what become ofthem, or if they ever did multi-page stuff
[11:58:15] *Werdna 	ponders perling it up
[11:58:24] <Duesentrieb>	i would have thought the wikibooks people would be more into this kind of thing...
[11:58:49] <Emily2>	I am a recent wikibook people
[11:58:59] <Duesentrieb>	Emily2: http://en.wikibooks.org/wiki/Wikibooks:PDF_Versions
[11:59:05] <Emily2>	this is our firs complete book
[11:59:13] <Duesentrieb>	Emily2: so, some people do it - ask them :)
[11:59:20] <emilsedgh>	where is the API of mediawiki stored? (or something like that so i find out how should i interact with db)
[11:59:25] <Emily2>	ok, thanks
[11:59:35] <Werdna>	emilsedgh: /api.php
[11:59:47] <Duesentrieb>	emilsedgh: api.php is the *web* api
[12:00:12] <Duesentrieb>	emilsedgh: what exactly are you looking for? what do you wnat to do, how, from where?
[12:00:24] <Duesentrieb>	!docs | emilsedgh
[12:00:24] --mwbot--	emilsedgh: An overview of available documentation about MediaWiki can be found at <http://www.mediawiki.org/wiki/Developer_hub>
[12:00:40] <emilsedgh>	Duesentrieb: i want to fetch a row from hook
[12:01:21] <Duesentrieb>	emilsedgh: then look at Database.php. the code is the best documentation. but you will find some docs linked from the developer hub. the doxygen stuff should get you a first overview.
[12:01:27] <emilsedgh>	i dont want to use stupid mysql_query() and mysql_fetch_row(), i want to learn and use native mediawiki api for that
[12:01:33] <Duesentrieb>	emilsedgh: anyway, look at how it's done elsewhere
[12:01:53] <Duesentrieb>	do, you shouldn't. there's a wrapper. it isn't very abstract, but do use it anyway
[12:02:28] <emilsedgh>	wfGetDB(DB_SLAVE)->getRow() is my answer?
[12:02:39] <TimStarling>	no
[12:02:42] *Werdna 	shall return.
[12:02:48] <TimStarling>	see docs/database.txt
[12:03:14] <Duesentrieb>	emilsedgh: nearly. not in one go. you get the db object using wfGetDB(DB_SLAVE), the you run your query, then you fetch rows.
[12:03:24] <Duesentrieb>	oh yea, listen to TimStarling
[12:03:34] <emilsedgh>	TimStarling: thanks :)
[12:03:40] <Duesentrieb>	TimStarling: i didn't even know that exists :P
[12:05:41] <peder>	just a question... i have a script generation users in a few wikis i have, but as don't fill in the uses-prefs field som users seem to get real weird defaults there... like editing boxes 4x4 in size and such
[12:06:26] <peder>	most get the 80x25 default i guess... but is there some prefs i definitly should define there?
[12:06:33] <Duesentrieb>	peder: maybe you should use an auth plugin
[12:06:40] <Duesentrieb>	!auth | peder
[12:06:40] --mwbot--	peder: http://www.mediawiki.org/wiki/AuthPlugin
[12:07:29] <Duesentrieb>	peder: alernatively, just read the prefs from a dummy user, copy them to a file, and populate the db field with that from your script
[12:07:45] <TimStarling>	here's my current problem, Werdna
[12:07:54] <TimStarling>	oh, you're gone
[12:08:46] <wikibugs>	03(NEW) Create namespaces for Hebrew Wikisource - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=13091 15enhancement; normal; Wikimedia: General/Unknown; (dovijacobs)
[12:08:53] <TimStarling>	anyway, my problem is that people are complaining that they see a read-only error with $1 as the reason, instead of the real reason
[12:09:16] <TimStarling>	this is because commitRollback returns a permission error of array(array('readonlytext'))
[12:09:53] <peder>	yeah, think that may be the most practical (just populating with some defaults)
[12:09:59] <TimStarling>	and then that precise return value is referenced three times like this:
[12:10:03] <TimStarling>			if( $result && $result !== array( array('readonlytext') ) ) {
[12:10:07] <TimStarling>					if( $error != array( 'readonlytext' ) ) {
[12:10:15] <TimStarling>			if( $result == array( array('readonlytext') ) ) {
[12:17:22] <TimStarling>	damnit
[12:17:26] <TimStarling>	that's not the problem
[12:19:12] <emilsedgh>	i hope this is the last question: i want to use scriptac and prototype in mediawiki, so that Headers are collapsed, when user clicks on the title, they expand.i know that i should add prototype and scriptac to my skin's header.but i dont know how should i change headers templating stuff.any idea where should i look? (is this right place for such questions?)
[12:20:03] <TimStarling>	that sounds like a job for pure javascript
[12:21:56] <emilsedgh>	TimStarling: scripat and prototype does the job, i could deal with that, just where could i define the html outputs for headers for example, so if a header is like print $title, i change it to somethins like print '<a href="javascript:void(0)" onclick="javascript:doExpand()">$title</a>'; (example)
[12:23:08] <Werdna>	TimStarling: I'm back
[12:24:12] <TimStarling>	I don't think it's the problem, I'm doing rollbacks in read-only mode and getting a normal error
[12:25:50] <TimStarling>	emilsedgh: Linker::makeHeadline()
[12:26:07] <emilsedgh>	TimStarling: thanks
[12:26:23] <emilsedgh>	'a course to mediawiki customization'
[12:27:51] <emilsedgh>	TimStarling: so it couldnt be overrided by skins? if no, what about a hook for it?
[12:28:11] <TimStarling>	Linker is the parent class of Skin
[12:28:22] <TimStarling>	so you can override it in the skin with a function of that name
[12:28:39] <emilsedgh>	ah yes
[12:28:40] <Seven-7>	Does mediawiki often bit rot?
[12:28:42] <emilsedgh>	'thanks'
[12:41:01] <emilsedgh>	and where are the contents below the headlines are defined?
[12:42:06] <Nikerabbit>	:o
[12:43:09] <emilsedgh>	<h$leve>something here></h$leve><p>some text</p>, where is that <p> defined, anyone knows?
[12:43:49] <emilsedgh>	s/$leve/$level
[12:44:24] <TimStarling>	that would be the rest of the parser
[12:45:34] <emilsedgh>	TimStarling: btw i really appreciate your help.thanks.
[12:51:57] <CIA-6>	03huji * r31148 10/trunk/phase3/languages/messages/MessagesFa.php: Adding/updating Persian translations
[13:04:50] <wikibugs>	03(mod) Replace the "magnify" symbol used below thumbnails by a more explicit one - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13070  (10huji.huji)
[13:08:23] <wikibugs>	03(mod) Pressing Enter in edit summary should not save the page - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13065  summary; +comment (10huji.huji)
[13:13:50] <wikibugs>	03(mod) Pressing Enter in edit summary should not save the page - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13065  (10magnusrk+wiki)
[13:14:25] <CIA-6>	03siebrand * r31149 10/trunk/extensions/GoogleMaps/ (EditorsMap.js GoogleMaps.i18n.php): Rename 3 more messages. 2 remaining, but it looks like they are unsed.
[13:17:57] <nakeee>	hey guys
[13:18:07] <nakeee>	how do I get mediawiki to look like wikibooks?
[13:18:10] <nakeee>	is it some extention?
[13:21:31] <TimStarling>	looks like plain monobook to me
[13:23:22] <nakeee>	I don't mean the skin I mean how the page look with the hirarchy
[13:24:26] <Duesentrieb>	nakeee: what hierarchy=
[13:24:34] <Duesentrieb>	link?
[13:24:56] <wikibugs>	03(mod) Special:LonelyPages shows transcluded pages - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=12764   +need-review +patch (10huji.huji)
[13:26:06] <nakeee>	http://en.wikibooks.org/wiki/A-level_Economics/AQA/Markets_and_Market_failure like that
[13:26:22] <nakeee>	yea the links:)
[13:26:36] <Duesentrieb>	you mean the table of contents at the top?
[13:26:47] <Duesentrieb>	this shows up automatically as soon as you have more than 3 headings on your page
[13:26:57] <Duesentrieb>	or more than four, i forget
[13:27:24] <nakeee>	no, I mean the fact that it puts chapters under the name of the books in the links
[13:27:34] <nakeee>	and it's easy to arrange the content
[13:28:02] <nakeee>	like A-level_Economics is the subject AQA is the name and Markets is the chapter
[13:28:13] <Duesentrieb>	"chapters under the name of the books in the links"? sorry, i donÄt get it
[13:28:27] <astro76>	nakeee: you just have to enable sub pages for the main namespace
[13:28:31] <Duesentrieb>	oh, you mean subpages
[13:28:33] <Duesentrieb>	!subpages
[13:28:33] --mwbot--	By default some namespaces have subpages enabled and others don't. You can enable or disable them using $wgNamespacesWithSubpages in LocalSettings.php. <http://www.mediawiki.org/wiki/Manual:%24wgNamespacesWithSubpages>
[13:31:02] <nakeee>	yea thanks
[13:31:05] <nakeee>	that what I wanted:)
[13:34:25] <CIA-6>	03siebrand * r31150 10/trunk/extensions/GoogleMaps/GoogleMaps.i18n.php: Rename the remaining two message anyway. Probably unused.
[13:35:10] <CIA-6>	03siebrand * r31151 10/trunk/extensions/Translate/Translate.php: Reactivate support for GoogleMaps. Message IDs have stabilised.
[13:36:47] <CIA-6>	03yaron * r31152 10/trunk/extensions/DataTransfer/languages/DT_Messages.php: Removed underscores from messages; adding underscores is now handled in the code
[13:38:31] <CIA-6>	03yaron * r31153 10/trunk/extensions/DataTransfer/includes/DT_GlobalFunctions.php: Fixed URL and description in extension data
[13:38:46] <CIA-6>	03yaron * r31154 10/trunk/extensions/DataTransfer/INSTALL: New version: 0.1.1
[13:53:29] <Sasa^Stefanovic>	off
[14:13:24] <CIA-6>	03siebrand * r31155 10/trunk/extensions/GoogleMaps/GoogleMaps.i18n.php: Fix find and replace errors
[14:25:01] <wikibugs>	03(mod) Replace the "magnify" symbol used below thumbnails by a more explicit one - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13070  (10guillom.pom)
[14:27:18] <alnokta>	how can you tell if mw is _really_ using eaccelerator cache? 
[14:28:27] <wikibugs>	03(NEW) New namespace request - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13092 normal; normal; Wikimedia: General/Unknown; (korkodush)
[14:30:50] <wikibugs>	03(mod) New namespace request - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13092  normal->15enhancement;  +shell; +comment (10raimond.spekking)
[14:31:33] <wikibugs>	03(mod) Create namespaces for Hebrew Wikisource - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13091  (10raimond.spekking)
[14:39:20] <CIA-6>	03siebrand * r31156 10/trunk/extensions/GoogleMaps/GoogleMaps.i18n.php: article -> page
[14:40:00] <alnokta>	?
[14:41:53] <wikibugs>	  - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=12242  +comment (10huji.huji)
[14:45:14] <steffkes>	hey all .. anyone who's tried this extension? http://www.mediawiki.org/wiki/Extension:PdfHandler .. it seems that it doesnt work !?
[14:45:34] <steffkes>	mediawiki isnt using the defined $wgMediaHandlers .. so, at no place, the handler is included !? *wired*
[14:48:08] <steffkes>	http://www.mediawiki.org/wiki/Manual:%24wgMediaHandlers .. say's that the var is still in use .. but an "grep" only shoes the file of this extension ..
[14:48:16] <steffkes>	no specialUpload Page or something like this
[14:52:03] <CIA-6>	03grondin * r31157 10/trunk/phase3/languages/messages/MessagesCs.php: Update localization message from BETAWIKI (core + aliases, skin names)
[14:56:08] <wikibugs>	03(mod) New:=?UTF-8?Q?=20Words=20beginning=20with?= =?UTF-8?Q?=20=D3=A9?=, =?UTF-8?Q?=20=D2=AF?=, =?UTF-8?Q?=20and=20=D1=91?= - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=13093  (10N/A)
[14:56:35] <wikibugs>	  - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13093  (10N/A)
[15:00:19] <wikibugs>	14(DUP)  - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13093  +comment (10Li-sung)
[15:00:29] <wikibugs>	03(mod) Support collation by a certain locale ( sorting order of characters) - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=164  +comment (10Li-sung)
[15:07:49] <[ag]ds>	anyone fielding API questions in here?
[15:08:07] <alnokta>	why this: php changePassword.php --user=someuser --password=somepass isn't working for me? even the sql query isn't working.. any idea why?
[15:08:13] <jlerner>	[ag]ds: just ask
[15:08:55] <[ag]ds>	oh, alright
[15:10:15] <[ag]ds>	http://en.wikipedia.org/w/api.php?action=query&list=search&srsearch=wikipedia&srlimit=10 <- the Query List Search API doesn't appear to work
[15:10:38] <[ag]ds>	When will this be implemented and is there another API call to a fulltext search that generates xml?
[15:11:02] <[ag]ds>	I'm trying to decide between MediaWiki and Twiki for enterprise use and this is a pretty important feature
[15:11:13] <xsa>	hi all
[15:11:41] <lymeca>	I run Debian Etch on my server with Apache 2.2, PHP 5.2.0, and MySQL 5.0.32
[15:11:46] <[ag]ds>	documentation page, http://www.mediawiki.org/wiki/Talk:API:Query_-_Lists#Fulltext_Search
[15:11:49] <alnokta>	hi xsa 
[15:11:56] <lymeca>	I want to know the best way to install MediaWiki 1.11.1
[15:12:08] <xsa>	rookie question, i'm trying to use the REVISIONID template, although it does not seem to work. I'm trying with [[Special:Statistics|{{REVISIONID}}]]. Am I missing anything?
[15:12:12] <lymeca>	I have it installed from Debian unstable but I cannot seem to get it to work yet.
[15:16:56] <[ag]ds>	any API gurus here?
[15:17:23] <alnokta>	lymeca) how did you install it from debian unstable?
[15:17:43] <lymeca>	apt-get build-dep mediawiki
[15:18:07] <lymeca>	apt-get -b source mediawiki
[15:18:28] <lymeca>	It installed dependencies and compiled it from source and created two .deb binary packages
[15:18:35] <lymeca>	I isntalled them using dpkg -i
[15:18:56] <alnokta>	never tried that
[15:19:26] <alnokta>	why not just get the file, unpack, create a db, use the installer?
[15:20:16] <lymeca>	get what file?
[15:20:34] <lymeca>	I'm actually migrating an existing mediawiki isntallation over to my server from another.
[15:20:47] <lymeca>	Apparently MediaWiki is looking at the incorrect place for the MySQL server
[15:21:13] <lymeca>	I updated config/LocalSettings.php with the correct MySQL info but the changes haven't taken effect yet.
[15:21:35] <lymeca>	How do I apply changes to LocalSettings.php?
[15:21:58] <alnokta>	you open it, you change what you want, you save it.
[15:22:07] <alnokta>	get the distribution file
[15:22:08] <xsa>	anybody knowledgeable in Templates and willing to help?
[15:22:33] <alnokta>	http://download.wikimedia.org/mediawiki/1.11/mediawiki-1.11.1.tar.gz
[15:25:24] <xsa>	do i need anything special to be able to use {{REVISIONID}} and friends?
[15:27:03] <alnokta>	no
[15:27:36] <[ag]ds>	alnokta have you used the API at all?
[15:27:48] <[ag]ds>	sorry to ask you directly but no one else seems responsive
[15:27:58] <alnokta>	unfortunately, no :)
[15:28:06] <xsa>	alnokta: coz either putting it raw, or using [[Special:Statistics:{{REVISIONID}}]] does not work.
[15:28:08] <[ag]ds>	ah, alrighty
[15:28:13] <alnokta>	no problem, help is asleep
[15:28:21] <[ag]ds>	i see
[15:28:38] <[ag]ds>	any time in the day when it picks back up?
[15:28:39] <wikibugs>	03(mod) Pressing Enter in edit summary should not save the page - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13065  (103tx38k)
[15:30:05] <[ag]ds>	alright, i'm going to just pop in later in the day. thanks alnokta
[15:33:28] <xsa>	:(
[15:34:29] <alnokta>	xsa) afaik, that instruction should never work
[15:34:55] <xsa>	alnokta: no it does not, but i cant find the right one.
[15:35:27] <xsa>	coz, for instance, [[Special:Statistics:{{CURRENTMONTH}}]] works
[15:35:36] <xsa>	so i tried with REVISIONID, but no go.
[15:35:54] <xsa>	not sure what i'm missing
[15:36:15] <alnokta>	oh that works? 
[15:36:38] <xsa>	ya, it does
[15:37:12] <xsa>	er, i mean, with replacing the : by |
[15:37:13] <xsa>	:)
[15:37:22] <xsa>	after Statistics, that is
[15:38:30] <alnokta>	yes 
[15:39:14] <xsa>	but all the {{REVISION*}} ones dont work
[15:43:40] <alnokta>	that is strange, it looks like it is enclosed between <nowiki /> tag
[15:44:47] <xsa>	oh, so you also see it?
[15:45:59] <wikibugs>	03(NEW) Incorrect Message with Pear Mail SMTP - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13094 minor; normal; MediaWiki: Email; (greg)
[16:04:22] <wikibugs>	03(mod) New namespace request - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=13092  (10korkodush)
[16:11:49] <wikibugs>	  - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=  +comment (10gangleri)
[16:21:07] <mmcgrath>	Is there any way to have a mediawiki page export to docbok?
[16:21:11] <mmcgrath>	err docbook
[16:24:57] <wikibugs>	03(mod) "Persistent" login expires many times within a single session - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=1396  +comment (10japoth)
[16:27:29] <CIA-6>	03siebrand * r31158 10/trunk/extensions/GoogleMaps/GoogleMaps.i18n.php: Fix typo (bg => br)
[16:29:47] <wikibugs>	03(mod) "Persistent" login expires many times within a single session - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=1396  (10japoth)
[16:29:51] <wikibugs>	03(NEW) Search by first letters or digits in [[Special:Categories] ] - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13095 15enhancement; normal; MediaWiki: Categories; (hegesippe46)
[16:31:59] <CIA-6>	03raymond * r31159 10/trunk/extensions/GoogleMaps/GoogleMaps.i18n.php: * 'cn' -> 'zh' but commented out, need clarification if 'zh-hans' or 'zh-hant'
[17:01:39] <Chuwiey>	Wiredtape is now officially changing his nick to Chuwiey :)
[17:07:41] <wikibugs>	03(NEW) Sidebar links - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13096 15enhancement; normal; Wikimedia: wikibugs; (valerij92)
[17:14:41] <CIA-6>	03raymond * r31160 10/trunk/ (10 files in 10 dirs): Localisation updates German
[17:30:25] <wikibugs>	14(INVALID) Sidebar links - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13096  +comment (10roan.kattouw)
[17:35:06] <mmcgrath>	Anyone know if mediawiki can export to docbook format?
[17:37:07] <Nikerabbit>	hello
[17:37:57] <alnokta>	no, it doesn't export in docbook format mmcgrath 
[17:38:01] <alnokta>	hi Nikerabbit 
[17:38:04] <Chuwiey>	hey Nikerabbit :)
[17:38:34] <Chuwiey>	Nikerabbit, is this one of those random hello's you've been having lately? :P
[17:39:21] <wikibugs>	03(NEW) Proposal for easy switching of language in Wikimedia pages - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13097 15enhancement; normal; Wikimedia: Bugzilla; (etjazz)
[17:39:23] <alnokta>	mmcgrath) that is mw export: http://www.mediawiki.org/w/index.php?title=Special:Export&action=submit
[17:41:30] <mmcgrath>	hmmm
[17:41:38] <wikibugs>	03(mod) Proposal for easy switching of language in Wikimedia pages - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=13097  (10etjazz)
[17:43:33] <jlerner>	is there any way not to turn [[Non-existent page]] into a link, for users that don't have edit privileges?
[17:44:43] <Chuwiey>	jlerner, this would take quite a bit of hacking of the parser, and then creating the permission check..
[17:44:49] <jlerner>	yeah
[17:48:18] <Nikerabbit>	Chuwiey_away: hurr
[18:13:03] <jeremyb>	mmcgrath: there are extensions for some wikis to render docbook content... (i know moin moin has for example)
[18:13:08] <jeremyb>	mmcgrath: whatchya doing?
[18:22:03] <CIA-6>	03rotem * r31161 10/trunk/phase3/languages/messages/MessagesHe.php: Update.
[18:29:36] <wikibugs>	03(FIXED) New namespace request - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=13092  +comment (10jeluf)
[18:34:48] <wikibugs>	03(FIXED) Create namespaces for Hebrew Wikisource - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=13091  +comment (10jeluf)
[18:37:13] <mmcgrath>	jeremyb: I'm actually looking to migrate away from moin to mediawiki
[18:37:33] <mmcgrath>	been working on a script (original: http://fedoraproject.org/wiki/  converted: https://publictest1.fedoraproject.org/wiki/ )
[18:37:46] <mmcgrath>	but the docs guys use the wiki as part of their workflow and exporting to docbook is part of that.
[18:46:05] <wikibugs>	03(FIXED) Modify some settings of Russian Wikibooks - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=12856  +comment (10jeluf)
[18:49:23] <wikibugs>	03(FIXED) Request to enable mswiktionary <- mswiki import - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13018  +comment (10jeluf)
[18:58:37] <wikibugs>	03(FIXED) Activating "patrol" on Italian Wiktionary - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13035  +comment (10jeluf)
[19:03:35] <wikibugs>	03(FIXED) Request for a Portal namespace on bg.wikipedia.org - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=13050  +comment (10jeluf)
[19:06:38] <wikibugs>	03(mod) Replace the "magnify" symbol used below thumbnails by a more explicit one - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13070  (10huji.huji)
[19:08:43] <wikibugs>	03(FIXED) Set Guarani wikipedia namespace name - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13060  +comment (10jeluf)
[19:09:49] <wikibugs>	03(FIXED) Add import source to de.wiki - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13089  +comment (10jeluf)
[19:12:18] <wikibugs>	  - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=12242  (10huji.huji)
[19:14:42] <wikibugs>	03(WONTFIX) Activate Extension:LanguageSelector on wmf wiki - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=12604  +comment (10jeluf)
[19:20:55] <CIA-6>	03greg * r31162 10/trunk/phase3/maintenance/tables.sql: Add wgDBTableOptions
[19:23:46] <wikibugs>	03(FIXED) Create mailing list for Wikimedia New York City group - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13080  +comment (10cbass)
[19:26:40] <Telemac>	Hello
[19:27:28] <silly>	Hello does anyone know how to set up disambiguation in mediawiki so that those pages are not placed into the main namespace?
[19:27:32] <Telemac>	Is there a way to prevent parser from adding "border" attribute when rendering [[Image:x]] ?
[19:28:21] <Raymond_>	Telemac: you mean the standard thumb design?
[19:28:34] <Raymond_>	Telemac: try [[image:x|frameless]]
[19:29:47] <jeremyb>	mmcgrath: example of typical page that would be sent to docbook?
[19:31:40] <Telemac>	Raymond_: Does not remove this attribute
[19:32:04] <Raymond_>	Telemac: :(
[19:32:58] <Chuwiey>	Telemac, you could make a css rule with higher specifity
[19:33:02] <Telemac>	This issue make page not XHTML strict
[19:33:21] <Telemac>	Chuwley: I don't mean about display, but about HTML source
[19:35:47] <Chuwiey>	Telemac, you won't be able to get xhtml strict anyway, I would hardly call mw output transitional if anything
[19:36:26] <Chuwiey>	unless you're planning on hacking the parser rather roughly that is.. and in that case you could just change the attribute
[19:37:10] <Telemac>	Excepted image it seems ok
[19:37:26] <Chuwiey>	do you have a link?
[19:37:45] <wikibugs>	03(mod) Sidebar links - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13096  (10valerij92)
[19:37:54] <mmcgrath>	jeremyb: here's one of our more complex example
[19:37:56] <mmcgrath>	http://fedoraproject.org/wiki/Docs/Beats?action=format&mimetype=xml/docbook
[19:38:16] <wikibugs>	03(mod) Sidebar links - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13096  (10valerij92)
[19:38:18] <Telemac>	http://www.nozicaa.com/en/
[19:40:02] <jeremyb>	wiki be slow
[19:40:33] <jeremyb>	or my internet maybe...
[19:41:10] <Chuwiey>	Telemac, first let me say, very nice :) ... second, this is a completely revised skin, so obviously it could be more xhtml strict compliant.. regarding images, you would just have to look at how they are rendered by the parser.. the border might even be a product of imagemagic.. 
[19:41:14] <mmcgrath>	jeremyb: nope, its the wiki.
[19:41:19] <mmcgrath>	jeremyb: now you know why we want to migrate :)
[19:41:24] <jeremyb>	hah
[19:41:33] <jeremyb>	i was doing a search and it finally gave 502
[19:41:37] <mmcgrath>	10,000 + pages on a flat filesystem (which is what moin uses) doesn't work that well.
[19:41:50] <jeremyb>	right
[19:42:04] <jeremyb>	they did have an SoC to move to abstracted storage
[19:42:07] <jeremyb>	and one for hg
[19:42:16] *jeremyb 	knows not what happened with them
[19:42:18] <mmcgrath>	they've had a lot of things
[19:42:19] <Telemac>	Chuwley: Yes a fully custom skin. I also think that I need to look at parser, I just hope there is a simplier way :)
[19:42:39] <Telemac>	Chuwley: Thanx
[19:42:49] <Chuwiey>	np, i didn't do anything :)
[19:43:03] *jeremyb 	wonders what the "remove spam" menu item does
[19:44:20] <Chuwiey>	I wonder if anyone knows of a way to make pdf->wikitext conversions.. as far as i'm concerned there's no need for tables and images, just text..
[19:44:49] <jeremyb>	and svg?
[19:44:49] <Chuwiey>	the only thing i've found so far is saving them as word documents or rtf and then converting that..
[19:44:59] <Chuwiey>	but that is horible as well..
[19:45:02] <mmcgrath>	jeremyb: any ideas on what to do there?  Its a blocker for us if we can't export to docbook, we don't have much internal php expertise.
[19:45:32] <Chuwiey>	mmcgrath, you could ask for an extension..
[19:45:54] <jeremyb>	mmcgrath: i'm comparing...
[19:45:58] <Chuwiey>	maybe someone who likes fedora will make it :) 
[19:46:00] <mmcgrath>	as in request someone write it?  whats the typical turnaround on something like that?
[19:46:07] *mmcgrath 	hopes so.
[19:46:12] <Chuwiey>	mmcgrath, i would say, unknown.
[19:46:14] <mmcgrath>	and I hope for your guys sake its not me ;-)
[19:46:23] *mmcgrath 	hasn't done php in years
[19:47:19] <Chuwiey>	mmcgrath, do you know of any program that can convert text to docbook? or wikitext for that matter? what about html?
[19:47:36] <Chuwiey>	if so, it could be fairly simple to access content via api and then use this program to convert..
[19:47:54] <wikibugs>	03(mod) Create namespaces for Hebrew Wikisource - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13091  (10dovijacobs)
[19:48:01] <mmcgrath>	I don't but that doesn't mean they don't exist.  I can certainly go that route.
[19:48:28] <Chuwiey>	i would think someone must have made a tool for html->docbook
[19:48:55] <mmcgrath>	I think they have one though the docs guys don't like to use it, it doesn't do a very good job I think.
[19:49:15] <Chuwiey>	here we go: http://www.xml.com/pub/p/423 (don't know if it works or anythiing like that :) )
[19:49:40] <Chuwiey>	only it's for purchase :-\
[19:49:44] <Chuwiey>	one sec..
[19:50:23] <Chuwiey>	here we go: www.eeecs.umich.edu./~ppadala/projects/tidy/
[19:50:48] <jeremyb>	grrrr
[19:50:53] <jeremyb>	*stab* verizon
[19:50:53] <mmcgrath>	server not found
[19:50:59] <Chuwiey>	which one?
[19:51:04] <mmcgrath>	www.eeecs.umich.edu./~ppadala/projects/tidy/
[19:51:10] <Chuwiey>	i'm looking at it..
[19:51:16] <Chuwiey>	oohh..
[19:51:26] <Chuwiey>	i wrote an extra e on eecs.. sorry
[19:51:37] <mmcgrath>	there we go, thanks.
[19:51:41] <Chuwiey>	np
[19:53:01] <Chuwiey>	mmcgrath, another thought is maybe wikimedia's library project has done something like this? might try asking there..
[19:55:17] <Chuwiey>	mmcgrath, here we go: this is what you probably should be looking at: http://meta.wikimedia.org/wiki/DocBook_XML_export
[19:55:20] <mmcgrath>	Do they have an irc presence?
[19:55:30] <Chuwiey>	unknown.. 
[19:56:16] <Chuwiey>	they even have a wikitext to docbook linux utility afaics.. 
[19:56:25] <Chuwiey>	and now i take my leave :) for lunch!
[19:57:07] <mmcgrath>	Thanks for that, I'll take a look at those, looks very promising.
[19:57:37] <Chuwiey_lunch>	np
[20:01:53] <wikibugs>	03(NEW) Article Page Edit. - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13098 15enhancement; normal; Wikimedia: General/Unknown; (valerij92)
[20:03:51] <wikibugs>	06(REMIND) Article Page Edit. - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13098  +comment (10valerij92)
[20:04:13] <CIA-6>	03greg * r31163 10/trunk/phase3/maintenance/ (4 files in 3 dirs): Add table page_props to Postgres schema.
[20:06:04] <CIA-6>	03greg * r31164 10/trunk/phase3/includes/LinksUpdate.php: Continue the pretty lining up from above.
[20:18:03] <CIA-6>	03siebrand * r31165 10/trunk/phase3/languages/messages/ (32 files): Localisation updates for core messages from Betawiki (2008-02-21 21:06 CET)
[20:20:32] <wikibugs>	04(REOPENED) Sidebar links - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13096  (10valerij92)
[20:21:00] <wikibugs>	04(REOPENED) Article Page Edit. - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13098  (10valerij92)
[20:21:55] <alnokta0>	siebrand) there?
[20:22:05] <siebrand>	alnokta0: somewhat
[20:22:17] <siebrand>	alnokta0: preparing extension commit.
[20:22:18] <alnokta0>	:)
[20:22:42] <alnokta0>	localisation ?
[20:23:27] *siebrand 	nods.
[20:23:45] <siebrand>	alnokta0: first time we have 100 i18n files in extensions changed...
[20:24:07] <alnokta0>	i was wondering, is it possible to get an email whenever there are new messages to translate or messages that became fuzzy? [on bw]
[20:24:30] <alnokta0>	siebrand) yes, there is progress :)
[20:24:47] <siebrand>	alnokta0: enable watchlist emails and put Betawiki:News on your watchlist
[20:25:50] <wikibugs>	03(mod) Sidebar links - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13096  +comment (10overlordq)
[20:26:02] <alnokta0>	siebrand) nice, thanks :)
[20:29:59] <alnokta0>	siebrand) why aln language is zero like that?
[20:30:35] <siebrand>	alnokta0: ?
[20:30:50] <alnokta0>	 aln 	 Gegë 	 0.00% 	 0.00% 	 0.00% 	 0.00%
[20:31:21] <siebrand>	alnokta0: that's because there are no localisations at all for it, just a fall back language.
[20:31:37] <siebrand>	sad stories all around...
[20:32:16] <alnokta0>	so it shouldn't be translated ? or nobody to do it?
[20:33:01] <siebrand>	alnokta0: I think the UI was added because of conditional approval at Wikimedia's Incubator.
[20:33:17] <siebrand>	alnokta0: and so far no one has knocked on the localisation door...
[20:33:35] <alnokta0>	:(
[20:36:25] <alnokta0>	now there is 84 langs with 70% localisation or more .. 
[20:36:27] <CIA-6>	03siebrand * r31166 10/trunk/extensions/ (98 files in 95 dirs): Localisation updates for extension messages from Betawiki (2008-02-21 21:06 CET)
[20:42:02] <alnokta0>	98 files :)
[20:47:25] <wikibugs>	14(INVALID) Sidebar links - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13096  (10roan.kattouw)
[20:48:31] <wikibugs>	03(mod) Article Page Edit. - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13098  +comment (10roan.kattouw)
[20:48:48] <wikibugs>	03(mod) Ad above article - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13098  summary (10roan.kattouw)
[20:52:22] <quentin_>	how can i create images with hyperlinks?
[20:53:31] <TimLaqua>	!e Icon
[20:53:31] --mwbot--	http://www.mediawiki.org/wiki/Extension:Icon
[20:56:15] <quentin_>	TimLaqua, it's the only why? no native syntax?
[20:56:23] <TimLaqua>	correct.
[21:05:19] <Splarka>	TimLaqua, quentin_: well, not exactly
[21:05:58] <Splarka>	you can enable http://www.mediawiki.org/wiki/Manual:$wgAllowExternalImages and then link them via <span class="plainlinks">[http://anylink http://anyimage]</span> for example
[21:07:36] <Chuwiey_lunch>	my god, "the office" is just absolutely insane.. :-P
[21:07:56] <Chuwiey_lunch>	or maybe it warrants a 8-P
[21:08:00] <quentin_>	Splarka, but this does not work with internal images, right? unfortunately you cannot set the size of an external image, as far as i know
[21:08:13] <Splarka>	well
[21:08:29] <Splarka>	an "external" image or link for a wiki just means absolutely defined
[21:08:53] <Splarka>	you can use local images and thumbnails thereof, if you know how
[21:09:23] <Splarka>	for example (if en.wp had external images enabled): http://en.wikipedia.org/wiki/Image:Wiki.png  is usually referred to via [[Image:Wiki.png]], but for an external image it would be  http://upload.wikimedia.org/wikipedia/en/b/bc/Wiki.png
[21:09:50] <Chuwiey_lunch>	quentin_ and also, though not a prefereable option, you could also just use redirect.. 
[21:09:50] <Splarka>	and for a 100px thumbnail it would be http://upload.wikimedia.org/wikipedia/en/thumb/b/bc/Wiki.png/100px-Wiki.png
[21:10:08] <Splarka>	Chuwiey_lunch: then you can only do it once per image, kinda annoying
[21:10:19] <Chuwiey_lunch>	hence, why it is isn't prefereable :)
[21:10:21] <alnokta0>	Splarka) how is the /b/bc/ is added?
[21:10:24] <TimLaqua>	indeed
[21:10:28] <Splarka>	md5
[21:10:43] <Chuwiey_lunch>	alnokta0, the first two digits of the md5 hash i believe..
[21:10:49] <Splarka>	alnokta0: hashed file paths, that is the first and first/second letter of the md5 of that file's filename
[21:11:03] <Chuwiey_lunch>	also, not all images will have thumbs of the specific size you want..
[21:11:06] <Splarka>	to evenly distribute the files among several directories, rather than have them all in one
[21:11:19] <Splarka>	Chuwiey_lunch: Wikimedia delivers thumbs on demand
[21:11:30] <Splarka>	for other installations, you have to generate and maintain the thumbs manually
[21:11:31] <Chuwiey_lunch>	Splarka, it does? 
[21:11:33] <alnokta0>	ah thx
[21:11:46] <Splarka>	so anyway
[21:12:00] <Splarka>	<span class="plainlinks">[http://en.wikipedia.org http://upload.wikimedia.org/wikipedia/en/thumb/b/bc/Wiki.png/100px-Wiki.png]</span>
[21:12:02] <Chuwiey_lunch>	Splarka, is this something that is generated for usual sizes on upload? or is this really "on-demand"/
[21:12:08] <alnokta0>	so if a file starts with v, what would be its hash? 
[21:12:17] <Splarka>	alnokta0: totally unpredictable
[21:12:25] <Splarka>	which is the point
[21:12:32] <quentin_>	Chuwiey_lunch, Splarka I just want to create a link like the "User","Devel", "Admin" images you have on mediawiki's homepage
[21:12:47] <Splarka>	if you did that, all the files that started with s would fill a folder and the z folder would be empty
[21:12:52] <quentin_>	or should i hardcode them?
[21:13:07] <Chuwiey_lunch>	quentin_ if you're only going to use them once, you can just use a redirect..
[21:13:16] <Splarka>	actually
[21:13:18] <Chuwiey_lunch>	probably would be the simplest method..
[21:13:20] <Splarka>	if you're only going to use them once
[21:13:22] <Splarka>	you can use CSS
[21:13:29] <Splarka>	MUCH simpler, and repeatable
[21:13:30] <alnokta0>	Splarka) so the hash for v in the first file could be different than it is in a second file?
[21:13:49] <Splarka>	$md5(v) = 9e3669d19b675bd57058fd4664205d2a
[21:13:57] <Splarka>	$md5(vv) = c4055e3a20b6b3af3d10590ea446ef6c
[21:14:01] <Chuwiey_lunch>	Splarka, background-image? afaik you can't use it inside wikitext..
[21:14:03] <quentin_>	how do you mean redirect?
[21:14:14] <Splarka>	Chuwiey_lunch: you can in site-wide
[21:14:18] <quentin_>	Splarka, and why css?
[21:14:25] <Splarka>	quentin_: don't use a redirect, use CSS, much cleaner
[21:14:26] <Chuwiey_lunch>	Splarka, true that's also a good option..
[21:14:41] <alnokta0>	nice long lines ;)
[21:14:43] <Splarka>	Chuwiey_lunch: I've done this many many times on Wikia for users ^_^
[21:15:09] <Chuwiey_lunch>	Splarka, however, I have encountered problems with this, especially for lower res users.. (where the image was too large) or when I needed the same image to be resized...
[21:15:30] <quentin_>	Splarka, i don't understand how you want to do it with CSS, you can only set background-images with css
[21:15:34] <Splarka>	Chuwiey_lunch: this is why you generate a thumbnail of it at the exact size you want, and define the size in CSS
[21:15:43] <Splarka>	quentin_: right, so you set that on a link
[21:15:45] <Splarka>	for example
[21:15:57] <Splarka>	<a class="imagelink_userhub">[[User hub]]</a>
[21:15:58] <Splarka>	er
[21:16:02] <Splarka>	<span class="imagelink_userhub">[[User hub]]</span> even
[21:16:07] <Splarka>	and then in Common.css"
[21:16:39] <Splarka>	.imagelink_userhub { width:100px; height:100px; background-image: url("http//yourimage")}
[21:16:40] <Chuwiey_lunch>	Splarka, not always an option.. for instance in my case i needed the image to be resized based on a template var..
[21:16:44] <Chuwiey_lunch>	but this is a very specific need
[21:16:44] <Splarka>	.imagelink_userhub a { width:100px; height:100px; background-image: url("http//yourimage")}   even
[21:16:58] <Splarka>	Chuwiey_lunch: well, that sounds like a personal problem ^_^
[21:17:07] <Chuwiey_lunch>	:)
[21:17:15] <Splarka>	but even then
[21:17:31] <Splarka>	you could just center the image and use a background color that merged in
[21:17:44] <Splarka>	or rely on browser scaling (icky compared to mediawiki thumbnailing)
[21:18:01] <Chuwiey_lunch>	i've just noticed that my nick when adding the "_lunch" becomes rather weird.. :)
[21:18:16] <Splarka>	quentin_: lemme knock up an example
[21:18:20] <Chuwiey_lunch>	Splarka, obviously it would depend on the image..
[21:18:41] <Splarka>	Chuwiey: lets help quentin_ first before going on a tangent
[21:18:48] <Chuwiey>	:)
[21:19:09] <Chuwiey>	quentin_, did you get the css examples Splarka gave?
[21:19:10] <Splarka>	alnokta0: http://en.wikipedia.org/wiki/MD5 read that
[21:19:36] <quentin_>	yes, I'm trying to apply this
[21:20:02] <Chuwiey>	btw, you can also do this on mediawiki:common.css or in your actual css files...
[21:20:43] <groovehunter>	hi. I'd like to ask if anyone knows some resources about approaches for integration of different markup syntax
[21:21:03] <Chuwiey>	groovehunter, be a bit more specific..
[21:21:19] <groovehunter>	not sure if it's worth for my use case to dive into the Parser code
[21:21:28] <Splarka>	Chuwiey: what do you mean "also"? where else can you do it?
[21:21:28] <groovehunter>	Chuwiey: ok, i try
[21:22:07] <Chuwiey>	Splarka, for instance, you might want to do it in main.css - if you felt like it..
[21:22:13] <Splarka>	ewwww
[21:22:26] <Chuwiey>	agreed. :)
[21:22:28] <Poundo>	after a user creates a new account they see this message "Could not send confirmation mail. Check address for invalid characters. " and sometimes the bit about "date.timezone setting" 
[21:22:46] <groovehunter>	My need is primary to display stx or restx and html pages (from filesystem) 
[21:22:50] <Poundo>	the account is createded however if this is a bug is there a work around
[21:22:55] <Splarka>	then it'd be skin specific
[21:23:17] <Chuwiey>	Poundo, where does this happen.. and what is the bit about date.timezone setting?
[21:23:42] <Chuwiey>	Splarka, true.
[21:23:49] <Poundo>	Chuwiey: it happens on the page returned after a new account is created
[21:23:52] <groovehunter>	my background is/was zwiki, and i find mediawiki parsing approach very different and up to now hard to understand
[21:24:05] <Chuwiey>	Poundo, i meant what wiki?
[21:24:08] <groovehunter>	(except for that it's php not python of course )
[21:24:38] <quentin_>	Splarka, unfortunately I do have difficulties with that workaround:
[21:24:45] <quentin_>	http://www.winusers.org/wiki/index.php/Main_Page
[21:24:46] <Splarka>	quentin_: http://test.wikipedia.org/wiki/Image_link
[21:24:53] <quentin_>	->softwareportal
[21:24:54] <Poundo>	Chuwiey: sorry my wiki ? the one I created is that what you mean
[21:25:01] <Chuwiey>	groovehunter, so are you looking for something that will allow you to use zwiki syntax, stx, restx and html in mediawiki? you might look into a converter of some sort..
[21:25:09] <Splarka>	here is what I did:
[21:25:10] <Splarka>	<span class="imagelink_wplogo">[[Wikipedia|&nbsp;]]</span>
[21:25:14] <Splarka>	.imagelink_wplogo a { width:100px; height:100px; background-image: url("http://upload.wikimedia.org/wikipedia/en/thumb/b/bc/Wiki.png/100px-Wiki.png"); display:block; text-decoration:none;}
[21:25:38] <Chuwiey>	Poundo, yes, what version, did this occur immediatly after install? and so on..
[21:25:45] <Poundo>	Chuwiey: this is the other message Mailer returned: It is not safe to rely on the system's timezone settings. Please use the date.timezone setting, the TZ environment variable or the date_default_timezone_set() function. In case you used any of those methods and you are still getting this warning, you most likely misspelled the timezone identifier. We selected 'America/Chicago' for '-6.0/no DST' instead 
[21:25:45] <Splarka>	although 115px height would be more accurate
[21:26:12] <Splarka>	quentin_: so make the link text &nbsp;
[21:26:26] <groovehunter>	My strong opinion is it would be useful if mediawiki could handle these markups on the fly so my guess was there are already some approaches to do so
[21:26:39] <Chuwiey>	Splarka, have you ever seen the error Poundo is referring to? 
[21:26:40] <Splarka>	and display:block;, then it will be the proper width and height
[21:26:57] <Chuwiey>	groovehunter, there might be, google might be helpful..
[21:26:57] <Splarka>	nope
[21:27:11] <Poundo>	 Chuwiey: Version is MediaWiki: 1.11.0 it continues well after I installed.
[21:27:39] <quentin_>	Splarka, display: block did it! ;)
[21:27:44] <Chuwiey>	Poundo, did it start immediatly after install or was something else the cause of it
[21:27:45] <Chuwiey>	?
[21:27:52] <groovehunter>	Chuwiey: yeah i collected a few. Just wanted to ask here also :)
[21:28:02] <Chuwiey>	groovehunter, :)
[21:28:29] <Splarka>	quentin_: for using resized images in link, just make an image of the size you want on a permanent page, like MediaWiki:Common.css/thumbnails (so it is protected)
[21:28:43] <Splarka>	...or upload them the proper size
[21:28:46] <groovehunter>	ok now a concrete question :)) CategoryTree itself displayed in sidebar...
[21:28:58] <Chuwiey>	groovehunter, ok.. 
[21:29:00] <groovehunter>	{{#CategoryTree:Root}}  
[21:29:17] <Poundo>	Chuwiey: Well i would have to say probably immediatly after install. It is happening on the initial install on the development box before I installed it on the production server and moved the php and localsettings files over
[21:29:19] <groovehunter>	and wikitext enabled in Monobook.php hack
[21:29:40] <groovehunter>	but i get the html source 
[21:30:04] <Chuwiey>	groovehunter, what do you mean by wikitext enabled in monobook.php hack? can't u just put the parser code in Mediawiki:Sidebar?
[21:30:13] <groovehunter>	http://www.mediawiki.org/wiki/Extension_talk:CategoryTree#CategoryTree_in_Sidebar_Clarification
[21:31:17] <Chuwiey>	Poundo, hmm.. I dunno, try svn version and see if you get the same error, this might be some sort of funky php error.. i don't really know.. 
[21:31:33] <groovehunter>	afaik one needs to enable wikitext rendering in sidebar to use it
[21:31:35] <groovehunter>	http://www.mediawiki.org/wiki/Extension:CategoryTree#Using_CategoryTree_in_the_Sidebar
[21:31:49] <Poundo>	Chuwiey: I should mention I am running on IIS6 on a windows server
[21:32:32] <Chuwiey>	Poundo, hmm.. which lends more to a problem not with the mw scripts but with you're php/iis configuration.. 
[21:32:50] <Chuwiey>	this might be a bug someone has mentioned before, try bugzilla...
[21:33:19] <Poundo>	Chuwiey: yea I am beginning to think it might be some configuration scramble
[21:33:44] <Poundo>	Chuwiey: bugzilla? whazz dat
[21:35:46] <Chuwiey>	groovehunter, from taking a look at the link, it might be that the error is caused by the replacement monobook.php hack, instead of using $wgOut they use echo there.. I don't know if it should work or not..
[21:36:01] <Chuwiey>	!bugzilla
[21:36:01] --mwbot--	All bugs in MediaWiki should be reported at http://bugzilla.wikimedia.org. This is also the place to request site configuration changes, new features or enhancements to existing features, although bear the following points in mind before making a feature/enhancement request: 1) If the request is specific to a Wikimedia wiki, please discuss the issue on that wiki first. 2) Consider whether a custom extension would be more appropriate.
[21:37:06] <Chuwiey>	and now i will go and rest for a bit before resuming another couple of hours of work..
[21:37:11] <Chuwiey>	:)
[21:37:27] <Poundo>	Chuwiey: Thanks for trying:-D
[21:38:26] <Chuwiey>	Poundo, np
[21:38:57] <wikibugs>	03(mod) Replace the "magnify" symbol used below thumbnails by a more explicit one - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13070  +comment (10dgerard)
[21:44:04] <gcardinal>	hi, what is the auto-edit tool for mediawiki? running on windows, auto 200 or something.
[21:46:19] <alnokta0>	brion) why not add wildcard dns for non-existing subdomains to redirect to wikipedia.org?
[21:46:44] <Poundo>	can I hack the Special:Userlogin so that the e-mail input box is not available for users creating an account?
[21:47:07] <alnokta0>	Splarka) thx , the subject is big ;)
[21:48:09] <brion>	alnokta0: there's some good reason for it, though i'm not sure what offhand
[21:48:19] <brion>	we used to use wildcard back in the day, it was changed to explicit list
[21:48:44] <groovehunter>	Chuwiey: could you explain more in detail please, where to use wgOut? Leave Monobook.php original?
[21:48:51] <flyingparchment>	alnokta0: wildcard CNAMEs are generally considered bad
[21:49:04] <psymin>	flyingparchment: Why is that?
[21:49:50] <alnokta0>	yes why 
[21:52:22] <sgtd>	Does the core Mediawiki provide an api to reuse/recheck the users authenticated-ness?  I'd like to have my unrelated PHP verify that the user is logged in to my mediawiki before doing x.
[21:58:28] <alnokta0>	i can see from google is that it can slow things down + not all servers implement it the same way..is that why?
[22:00:22] <flyingparchment>	we used to have a wildcard for *.wikimedia.org.  every time someone mistyped a server name, it would connect to some other server and produce confusing and non-obvious errors
[22:00:52] <flyingparchment>	when you have a wildcard, you can't check for existence without polluting the cache and then you have to wait for it to expire before you can actually add the record (or find everywhere it's cached and flush it)
[22:01:33] <gcardinal>	how to export pages based on the category ?
[22:02:29] <Splarka>	gcardinal: the Special:Export in trunk has an "Add pages from category" box
[22:02:31] <Splarka>	http://en.wikipedia.org/wiki/Special:Export
[22:02:45] <Duesentrieb>	has been there for i while. i think i added that in 1.10
[22:02:55] <Duesentrieb>	or was it 1.11?
[22:03:42] <alnokta0>	flyingparchment) you mean the dns server checking for subd existance..right?
[22:03:48] <BrokenArrow>	can somebody double-check the categorymembers ordering in the api? I have a bot operator complaining that since a few days it looks broken
[22:04:02] <flyingparchment>	alnokta0: i don't understand the question
[22:04:07] <BrokenArrow>	(ehr... please, of course :-))
[22:04:28] <alnokta0>	flyingparchment) "you can't check for existence without polluting the cache and then you have to wait for it to"
[22:04:45] <flyingparchment>	alnokta0: i mean me typing 'dig whatever.wikipedia.org' at the command line to see whether it exists
[22:04:54] <Poundo>	anyone have any tips on how to hack Special:Userlogin i want to get rid of the email input. Thats the only way I have found to eliminate the "Could not send confirmation mail. Check address for invalid characters. " notice
[22:05:02] <Poundo>	not email no notice!
[22:05:50] <Poundo>	no email , no notice!
[22:07:10] <sgtd>	Duesentrieb: Does the core Mediawiki provide an api to reuse/recheck the users authenticated-ness?  I'd like to have my unrelated PHP verify that the user is logged in to my mediawiki before doing x.
[22:07:44] <alnokta0>	ah
[22:08:31] <Duesentrieb>	sgtd: i don't know, sorry. 
[22:08:54] <Atluxity>	sgtd; api.php
[22:08:58] <Atluxity>	2 sec
[22:09:10] <Duesentrieb>	there's an openid plugin, maybe look at that
[22:09:21] *sgtd 	goes off to read api.php
[22:09:34] <Duesentrieb>	Atluxity: well, api.php can check if you are logged in. how would another webapp check that you are logged in?
[22:09:39] <Duesentrieb>	shared cookies?
[22:10:00] <Duesentrieb>	if it's on the same domain, that might work...
[22:10:37] <sgtd>	i read api.php as an alternate method for mediawiki to authenticate users.
[22:10:43] <Atluxity>	sgtd: api.php?action=query&meta=userinfo was what i was thinking about
[22:10:57] <Atluxity>	Duesentrieb: you are right
[22:11:05] <Atluxity>	I dont know how that would be done
[22:11:09] <Atluxity>	in a secure way
[22:11:48] <sgtd>	there's no  include(../includes/User.php)     if($isLoggedIn)...?
[22:11:55] <Duesentrieb>	i'm pretty sure that the api's functionality can be used to get this to work somehow, if you are an the same domain and can share cookies. i'd have to think about how exactly, though. and i don't think we have something ready-to-used
[22:11:59] <sgtd>	it is PHP running on the same server in the same domain
[22:12:23] <Atluxity>	openid plugin would be a good track, in my opinion
[22:12:35] <Duesentrieb>	sgtd: something like that could be written, i supposed, but i donÄt think it exists.
[22:12:45] <badspyro>	anybody know where the default directory for mediawiki is in debian?
[22:12:51] <sgtd>	gotcha.  thanks Duesentrieb  and Atluxity!
[22:13:09] <Duesentrieb>	Atluxity: does openid offer session support, i.e. single login? isn't it an authentication service, rather?
[22:13:20] <gcardinal>	if I have 2000 xml files in dir /tmp/ how can I import all of them in one go using importDump.php ?
[22:13:44] <Duesentrieb>	gcardinal: for f in /tmp/*.xml; do php importDump.php $f; done
[22:14:27] <Duesentrieb>	yes, "done" is part of the command
[22:14:41] <flyingparchment>	for f in /tmp/*.xml; php importDump.php $f
[22:14:43] <flyingparchment>	;)
[22:15:06] <flyingparchment>	(doesn't everyone use zsh?)
[22:15:12] <Atluxity>	Duesentrieb: uncertain
[22:15:19] <Duesentrieb>	Atluxity: me too
[22:16:03] <gcardinal>	wow nice :) thx
[22:16:22] <badspyro>	anybody know where the mediawiki config is?
[22:16:54] <quentin_>	is it possible to easily share the same image folder and the same user db with interwiki?
[22:18:00] <gcardinal>	badspyro - mediawiki/config/LocalSettings.php OR mediawiki/LocalSettings.php
[22:18:23] <badspyro>	thanks
[22:20:15] <|gatsby|>	Greetings. Anybody know if the mediawiki search page _form_ is customisable?
[22:20:31] <|gatsby|>	I don't see any appropriate hooks in the hook list.
[22:20:54] <Splarka>	quentin_: you mean, shared user db across domains?
[22:20:58] <Chuwiey>	|gatsby|, everything is customizable, it depends on the degree of hacking you want to do.. :
[22:21:01] <Chuwiey>	:)
[22:22:02] <quentin_>	Splarka, well, I have two wikis installed (for different languages) and would like to share the same users (db) and same folders (skin, images)
[22:22:18] <Splarka>	http://www.mediawiki.org/wiki/Manual:$wgSharedDB   ... I know Wikia uses the same user DB across multiple domains (although you are not automatically logged in to all of them... logging in to www.wikia.com logs you in to doom.wikia.com but not memory-alpha.org, as the cookies can't cross that root)
[22:22:19] <|gatsby|>	Chuwiey: well, I'd like to avoid hacks if possible and get this done cleanly :)
[22:22:25] <Splarka>	they also share skins yah
[22:23:49] <sgtd>	Duesentrieb / Atluxity: Ooo! I spy "function isLoggedIn()" in ../includes/User.php
[22:23:52] <Splarka>	(I suppose there is a way to have a dedicated cookie domain, like cookies.wikia.com, and then authentication would involve an image or iframe call to that server which would log you in to all the other domains, or something equally silly... but there hasn't been a big demand for it on Wikia)
[22:24:02] <Chuwiey>	|gatsby|, i don't think there is a hook for it, no.. o>=<
[22:24:02] <quentin_>	Splarka, how can I config my wikis to have common folders (skins)
[22:24:10] <Splarka>	no idea, ask #wikia ^_^
[22:24:17] <|gatsby|>	Chuwiey: The function in question appears to be powerSearchbox
[22:24:39] <Jack_Phoenix>	Splarka: hai. also, in case if you forgot, wikia is not open source, we at #wikia don't know anything :P
[22:24:45] <|gatsby|>	Chuwiey: what is the recommended practice for such situations? Direct editing?
[22:24:54] <Duesentrieb>	sgtd: sure that exists. but User.php will not work standalone. it relies upone being used inside the mediawiki application.
[22:25:01] <Splarka>	Jack: yah but you're just a user (with higher access, but still a user)
[22:25:08] <Duesentrieb>	sgtd: if you manage to disentangle it from the rest of the system, you're all good.
[22:25:08] <Splarka>	no disrespect... but no respect either ^_^
[22:25:41] <Jack_Phoenix>	meh. still, #mediawiki knows more about how MediaWiki works than Wikia
[22:25:48] <Splarka>	actually...
[22:25:58] <sgtd>	Duesentrieb: hmmm...  sounds like work, but doable.
[22:26:15] <Chuwiey>	|gatsby|, well, it really depends on what you want to accomplish, because there have been talks of making a better search interface (specifically the ajax auto suggest) though, that doesn't seem relevant to what you want to do...
[22:26:17] <Duesentrieb>	sgtd: indeed. if you manage, please share on mediawiki.org :)
[22:26:19] <Splarka>	what you could maybe do is set the /skins path to a rewrite rule... a 302 or symbolic link
[22:26:52] <flyingparchment>	spam subject: "you have nothing to lose, just a lot to gain!" ... i guess, this is an advert for an ineffective diet pill
[22:27:17] <Splarka>	"magic pill to move all the fat in your gut into your penis"
[22:27:23] <|gatsby|>	Chuwiey: I'm looking to ditch all the name space fields and also add an option to search and linked Drupal installation as well (via a redirect).
[22:27:34] <|gatsby|>	s/and/an
[22:28:14] <Chuwiey>	Splarka, where are you getting that from?
[22:28:45] <Splarka>	the penis fat or the rewrite advice?
[22:28:54] <Jack_Phoenix>	:P
[22:29:04] <Chuwiey>	|gatsby|, hmm.. disabling all the namespaces can be done from localsettings iirc (though i don't know if it makes them disappear on the form) .. regarding drupal, i would think this needs a hack of the code...
[22:29:14] <Chuwiey>	Splarka, the penis fat :P
[22:29:26] <Splarka>	just seems like the perfect spam hook
[22:29:32] <Chuwiey>	or maybe that's connected to the rewrite advice?
[22:29:59] <Chuwiey>	don't tell me you just thought of that on the fly ;)
[22:30:12] <Splarka>	"cheap aluminum siding for your house to hide the fact that you watch our free porn and take our cheap meds while saving money on your car insurance!"
[22:30:24] <Splarka>	I did, I'm bad
[22:31:31] <Chuwiey>	go work in marketing :)
[22:32:06] <Skizzerz>	"Dehydrated H20 -- just add water!"
[22:32:25] *Jack_Phoenix 	buys that
[22:32:49] <wikibugs>	03(mod) Replace the "magnify" symbol used below thumbnails by a more explicit one - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13070  (10huji.huji)
[22:32:51] <|gatsby|>	Chuwiey: Cheers
[22:33:07] <Chuwiey>	|gatsby|, np :)
[22:33:20] <Splarka>	Chuwiey: hmm, your image redirecting trick doesn't seem to work on wikimedia wikis anymore
[22:33:34] *|gatsby| 	is now thinking of doing this via JS and be done with it :)
[22:33:42] <Chuwiey>	Skizzerz, i wonder if h20 is a better version of h2o? :)
[22:33:54] *Splarka 	thinks certain inter-namespace redirects have been broken since 1.7 or so
[22:33:58] <Chuwiey>	Splarka, probably because of image licensing problems?
[22:34:20] <Splarka>	probably because it was a horrible hack ^_^
[22:34:26] <Chuwiey>	:)
[22:35:06] <Splarka>	here, I started a help page draft: http://www.mediawiki.org/wiki/User:Splarka/Help:Linked_images
[22:35:18] <Jack_Phoenix>	Spl: now go start your userpage :>
[22:35:33] <quentin_>	Splarka, what does this mean: "The databases needs to have matching (or no) table prefixes in order to work"
[22:36:12] <quentin_>	I have installed both wikis into the same db with different prefixes: no prefix for the english version, and de_ for the german
[22:36:17] <quentin_>	is it okay?
[22:36:33] <|gatsby|>	does MW use any JS libraries?
[22:37:13] <|gatsby|>	or is there a jquery extension by any chance?
[22:37:15] <Duesentrieb>	afaik it's all pretty much home-bew
[22:37:18] <Splarka>	|gatsby|: just its own, in wikibits.js 
[22:37:34] <wikibugs>	03(mod) Replace the "magnify" symbol used below thumbnails by a more explicit one - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13070  (10dgerard)
[22:37:36] <Splarka>	and a few specific ones, such as on the blocking page, and ajax watching, etc
[22:37:38] <Duesentrieb>	|gatsby|: you can embed any js libs easily enough
[22:37:47] <Duesentrieb>	|gatsby|: just edit MediaWiki:common.js
[22:37:54] <|gatsby|>	Splarka, Duesentrieb: Cheers
[22:38:02] <Splarka>	quentin_: where is that documentation?
[22:38:06] <|gatsby|>	I'll look into that, thanks
[22:38:11] <Duesentrieb>	|gatsby|: that's a *page* btw, not a file
[22:38:18] <|gatsby|>	Gotcha :)
[22:38:35] <quentin_>	Splarka, on the link you just sent to me!
[22:38:38] <Splarka>	Jack: sofixit
[22:38:53] *Jack_Phoenix 	lazes
[22:39:25] <Splarka>	Jack: you're a sysop right? can you add the CSS example to the common.css on mw? ^_^
[22:39:39] <Jack_Phoenix>	ya, sure thing
[22:39:44] <Splarka>	quentin_: I don't know how it is done, I just know it is done, I'm a user-side user
[22:39:45] <wikibugs>	03(mod) Replace the "magnify" symbol used below thumbnails by a more explicit one - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13070  +comment (10cbass)
[22:39:49] <Jack_Phoenix>	and you should be a MW dev, Spl ;)
[22:39:59] *Skizzerz 	prepares a rollback of Jack_Phoenix's edit just for the fun of it
[22:40:09] <Splarka>	Jack: PHP and MySQL boggle the hell out of me
[22:40:27] <Splarka>	I like to do things from the user POV
[22:40:28] <Jack_Phoenix>	so? they probably need JS gurus too ;-)
[22:40:32] <Jack_Phoenix>	+CSS too
[22:40:32] <Splarka>	heh
[22:40:46] <Splarka>	riight, the one CSS patch I advised broke opera
[22:41:04] <Splarka>	td + td in monobook/main.css caused opera 9 to slow down insanely, even on pages the rule wasn't applied
[22:41:09] <Jack_Phoenix>	opera? bah, who needs them!
[22:41:51] <Jack_Phoenix>	Spl: the .imagelink_wikilogo a CSS example from your help page, right?
[22:42:01] <Splarka>	right, but verify I didn't screw it up
[22:42:07] <wikibugs>	03(mod) Replace the "magnify" symbol used below thumbnails by a more explicit one - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13070  (10dgerard)
[22:43:07] <Jack_Phoenix>	you know better than I do ;-)
[22:43:16] <Jack_Phoenix>	http://www.mediawiki.org/wiki/MediaWiki:Common.css - 'tis live now
[22:43:28] <Splarka>	but you'll get the blame, no consequences for me
[22:44:15] <Skizzerz>	Splarka: I think putting redirects on image pages is still possible
[22:44:50] <Splarka>	Skizzerz: http://test.wikipedia.org/wiki/Image_talk:Shim.svg
[22:45:19] <Splarka>	perhaps a config setting (I couldn't find it though)
[22:45:25] <Skizzerz>	hmm... perhaps because it's an SVG?
[22:45:27] <Splarka>	unless it is the one for hard-redirects
[22:45:33] <Skizzerz>	idk
[22:45:39] <Skizzerz>	It works on 1.11
[22:46:41] *Splarka 	hard reloads common.css
[22:47:27] <Splarka>	Jack: ta, works!
[22:48:15] <Chuwiey>	has anyone experience with semanticforms?
[22:48:22] <Jack_Phoenix>	np :)
[22:53:42] <wikibugs>	03(mod) Replace the "magnify" symbol used below thumbnails by a more explicit one - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=13070  (10huji.huji)
[22:54:55] <Atluxity>	how can I take part in the development of api.php, ask questions and request features?
[22:55:09] <Splarka>	!bugzilla
[22:55:09] --mwbot--	All bugs in MediaWiki should be reported at https://bugzilla.wikimedia.org. This is also the place to request site configuration changes, new features or enhancements to existing features, although bear the following points in mind before making a feature/enhancement request: 1) If the request is specific to a Wikimedia wiki, please discuss the issue on that wiki first. 2) Consider whether a custom extension would be more appropriate.
[22:59:28] <Splarka>	Atluxity: so for example,  here are all the API bugs/feature requests on bugzilla:  https://bugzilla.wikimedia.org/buglist.cgi?query_format=advanced&short_desc_type=allwordssubstr&component=API   
[23:01:56] <Atluxity>	Yeah, that looks like a good entrypoint to learning more
[23:01:57] <Atluxity>	thanks