[00:10:40] <CIA-13>	03david * r24292 10/branches/liquidthreads/ (7 files in 4 dirs): Talkpage headers are now feature-complete.
[00:15:13] <TimStarling>	concurrent cleanup of a remote content-hashed file archive turns out to be quite a headache
[00:16:26] <robchurch>	"There was, um, an error. And we lost all images and media. All of it. No backups. Dog ate them."
[00:16:29] <TimStarling>	you could lock all the index gaps and then delete the files in a batch, but that could give a lot of lock contention and it would probably lead to deadlocks
[00:17:29] <TimStarling>	you could lock each content hash one at a time and delete the file while it is locked, but in the current remote model using PHP, that would be very slow
[00:18:35] <TimStarling>	the options appear to be: 1) lock everything and cross fingers that the contention is not too bad or 2) make the second method not slow
[00:19:03] <TimStarling>	how come brion is away? lazy bastard ;)
[00:19:09] <robchurch>	heh
[00:21:06] <TimStarling>	the current method is to use a global lock, serialising all file archive operations
[00:21:27] <TimStarling>	so locking all the rows in question couldn't be worse than that
[00:21:37] <TimStarling>	except the deadlock problem
[00:23:39] <TimStarling>	there's a variant of (2) which would be quite cool: use a daemon which does both file and DB operations
[00:24:03] <TimStarling>	the daemon would be on a single host
[00:24:18] <TimStarling>	so cleanup would be globally serialized, but you could also merge duplicate requests
[00:24:35] <TimStarling>	by far the hardest way to do it though
[00:25:32] <robchurch>	We know you like a challenge, Tim.
[00:30:10] <TimStarling>	no, I think the way to do it is to have the storage host contact the DB server
[00:37:51] <Avatar_of_Time>	good night^^
[00:44:47] <CIA-13>	03robchurch * r24293 10/trunk/phase3/skins/common/images/spinner.gif: Remove green blob from the centre (was supposed to be part of the "scaled-down flower" concept, but just looks odd at this size)
[00:53:47] <Leefmc>	im trying to setup a test site, and its under mysite.com/mw/mirror. The problem is i think i can't get the scriptpath right
[00:54:05] <Leefmc>	is "$wgScriptPath       = "/mw/mirror";" correct?
[01:00:59] <CIA-13>	03robchurch * r24294 10/trunk/phase3/ (RELEASE-NOTES docs/hooks.txt includes/User.php): Introduce 'UserGetRights' hook; see docs/hooks.txt for more information
[01:01:15] <CIA-13>	03robchurch * r24295 10/trunk/extensions/AutomaticGroups/ (AutomaticGroups.php README): Support setting rights as well
[01:08:34] <CIA-13>	03robchurch * r24296 10/trunk/extensions/SyntaxHighlight_GeSHi/SyntaxHighlight_GeSHi.i18n.php: Estonian localisation (c/o et:Kasutaja:M2s17)
[01:25:30] <CIA-13>	03robchurch * r24297 10/trunk/extensions/SyntaxHighlight_GeSHi/SyntaxHighlight_GeSHi.i18n.php: Trim whitespace
[01:26:55] <CIA-13>	03robchurch * r24298 10/trunk/extensions/SyntaxHighlight_GeSHi/ (SyntaxHighlight_GeSHi.class.php SyntaxHighlight_GeSHi.php): Honour 'LoadAllMessages' hook; make sure messages show up in Special:Allmessages, etc.
[01:58:35] *Lyma 	off
[02:34:03] <CIA-13>	03david * r24299 10/branches/liquidthreads/extensions/LqtModel.php: trivial typo
[02:41:10] <Leefmc>	hmm
[02:41:37] <CIA-13>	03david * r24300 10/branches/liquidthreads/extensions/LqtExtension.php: link from headers back to talk pages
[02:55:35] <Leefmc>	is it dangerous to have 2 wiki's use the same DB? (one testing, the other live)
[02:55:40] <Leefmc>	well n/m
[02:55:41] <Leefmc>	lol
[02:56:25] <^demon>	Leefmc: Eh, not really.
[02:56:36] <Leefmc>	oh wow, k kool
[02:56:42] <^demon>	Are they using the same prefix for table names?
[02:56:47] <Leefmc>	yup
[02:56:55] <Leefmc>	its a mirror
[02:57:05] <^demon>	Well, the only issue would be editing articles on the test side...because those changes would go on the live side too.
[02:57:11] <Leefmc>	yea
[02:57:19] <^demon>	But if you want to do it just so you can test software features and new extensions, you're ok.
[02:57:26] <Leefmc>	k cool
[02:57:49] <Leefmc>	its weird though, i think an extension, or SOMETHING, i added is causing timeout issues and index.php to load improperly
[02:58:43] <Leefmc>	i've got a fresh wiki installed, with just the mirror DB being used, and it performs great. My other version however, well in firefox only (go figure) it makes me download index.php every 2-10 page loads
[02:59:03] <sj>	is there a way to change th default skin based on browsertype?
[02:59:18] <sj>	one for ns 1.0, one for ff 3, one for ie...
[02:59:39] <^demon>	sj: You might be able to do it via JS.
[02:59:40] <Leefmc>	if you know php i'd imagine it would be easy
[02:59:46] <Leefmc>	or JS, yea
[03:00:25] <sj>	I don't know how to call out to browsertype in php.  else I assume I could put a switch around the right lines in localsettings ...
[03:00:34] <^demon>	Call their user agent.
[03:00:42] <^demon>	$_SERVER['USER_AGENT'];
[03:01:59] <sj>	ah
[03:02:04] <sj>	thx
[03:03:21] <^demon>	No worries.
[03:05:33] <Leefmc>	can users change their passwords?
[03:06:18] <^demon>	Leefmc: Yes.
[03:06:47] <Leefmc>	hmm sry bout that, lot on my mind. Come to think of it its right in Prefs
[03:06:54] <Leefmc>	heh, ty for not slapping such a stupid question :)
[03:24:49] <robchurch>	No, users all have to use the password "foobaz", because a bug report filed six months ago indicated that they were otherwise difficult to crack. For that matter, sysop passwords are all "barbat".
[03:28:22] <wikibugs>	03(NEW) Prevent uploads with missing descriptions - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=10650 15enhancement; normal; MediaWiki: Uploading; (dragons_flight)
[03:30:40] <davidmccabe>	How can I make sure I am seeing the very latest up-to-date information in the database?
[03:30:58] <^demon>	davidmccabe: Put ?action=purge at the end of a page.
[03:33:22] <davidmccabe>	Specifically, I want to move a page, and then, within the same PHP instance, be able to look up the page's title and see the new title.
[03:33:27] <davidmccabe>	in order to do a cascading move.
[03:33:46] <davidmccabe>	action=purge calls Title::invalidateCache(); I don't see anything happening there that doesn't already happen when a page is moved.
[03:34:11] <wikibugs>	03(NEW) Add a source field to the upload form - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=10651 15enhancement; normal; MediaWiki: Uploading; (dragons_flight)
[03:35:16] <^demon>	If moving already calls invalidateCache(), then you are seeing the most up-to-date information.
[03:35:56] <davidmccabe>	Maybe there's some magic there I'm not spotting; I'll try it.
[03:37:02] *davidmccabe 	shakes his head.
[03:37:21] <davidmccabe>	Sometimes life is just funny. I write this feature, it doesn't work at all for the reason alluded to above,
[03:37:23] <davidmccabe>	I leave it alone for a month.
[03:37:28] <davidmccabe>	Come back to fix it, and it works immediately.
[03:39:17] <^demon>	Glad it works.
[03:39:23] <davidmccabe>	thanks.
[03:39:35] <CIA-13>	03david * r24301 10/branches/liquidthreads/extensions/LqtExtension.php: Reasonable behavior if there are no threads in the archive. Also, turned subject-renaming back on to start debugging it, but now it just works without any changes. No idea why.
[03:39:40] <davidmccabe>	I guess somebody changed something in MW in the meantime that affects it.
[03:39:47] <davidmccabe>	oh, CIA is back.
[03:42:54] <CIA-13>	03david * r24302 10/branches/liquidthreads/ (10 files in 7 dirs): (log message trimmed)
[03:42:56] <CIA-13>	Merged revisions 24276-24301 via svnmerge from
[03:42:58] <CIA-13>	svn+ssh://david@svn.wikimedia.org/svnroot/mediawiki/trunk/phase3
[03:43:00] <CIA-13>	........
[03:43:02] <CIA-13>	 r24277 | avar | 2007-07-19 17:05:45 -0700 (Thu, 19 Jul 2007) | 2 lines
[03:43:04] <CIA-13>	 Report the lines win32 newlines are found at
[03:43:18] <CIA-13>	........
[03:43:49] <robchurch>	davidmccabe: If you want to make sure you are seeing the absolute latest data, use the master.
[03:44:15] <davidmccabe>	robchurch: was using the master all along.
[03:44:16] <robchurch>	This is far easier said than done, especially if you then want to call functions where the database access is abstracted away. :(
[03:44:21] <davidmccabe>	anyways it seems to work now. no idea why it didn't before.
[03:44:34] <robchurch>	Oh, good.
[03:44:35] <davidmccabe>	robchurch: it's not like such functions come up very often.
[03:55:05] <CIA-13>	03aaron * r24303 10/trunk/extensions/FlaggedRevs/ (FlaggedRevs.php FlaggedRevsPage_body.php): *OOP stuff
[04:00:12] <qsheets>	YAY
[04:00:23] *qsheets 	finally did it
[04:00:26] <qsheets>	...
[04:00:38] <davidmccabe>	did what qsheets?
[04:00:55] <qsheets_>	look
[04:01:16] <qsheets_>	I linked wiki users to irc nicks
[04:01:50] *davidmccabe 	doesn't get it.
[04:02:20] <qsheets_>	one sec, and i'll post a picture
[04:03:53] <robchurch>	VoiceOfAll: You don't need to pass objects by reference any more.
[04:04:43] <robchurch>	(Or more accurately, you don't need to explicitly etc.)
[04:06:12] <VoiceOfAll>	robchurch: so if I pass $s, it is no longer a local clone?
[04:06:26] <robchurch>	Not if it's an object, no.
[04:06:32] <VoiceOfAll>	wouldn't that break some sloppy stuff
[04:06:35] <VoiceOfAll>	?
[04:06:41] <robchurch>	I'm talking about your new code.
[04:07:09] <davidmccabe>	so are objects always pass-by-reference, then, or only when the receiving method marks them as such?
[04:07:20] <qsheets_>	hmph
[04:07:20] <robchurch>	Previously, the latter; that is, before PHP 5.
[04:07:29] <robchurch>	PHP 5 corrected this insane behaviour.
[04:07:33] <VoiceOfAll>	yes, but I'm just wondering about old functions that edited them for process purposes, not to actually edit the original copy
[04:08:12] <davidmccabe>	robchurch: so what you're saying is ... :)   in PHP 5, all objects are always pass-by-reference.
[04:08:24] <robchurch>	Well, that's a good question. With some luck, we found and fixed up all those bits of code to use clones where essential.
[04:08:31] <robchurch>	But I think we just got lucky there. ;)
[04:08:36] <VoiceOfAll>	right
[04:08:38] <VoiceOfAll>	:D
[04:08:58] <qsheets_>	davidmccabe: look at http://mwtesting.gotdns.com/prefs.png and http://mwtesting.gotdns.com/chat.png
[04:09:19] <VoiceOfAll>	robchurch: OK, well I still have an ampersand fetish
[04:09:32] <qsheets_>	mainly prefs
[04:09:33] <VoiceOfAll>	they look red and festive in my php editor
[04:09:49] <davidmccabe>	qsheets_: You are integrating IRC into MediaWiki. We should talk.
[04:09:58] <qsheets_>	lol
[04:10:07] <davidmccabe>	Is that thing javascript or flash or what?
[04:10:13] <qsheets_>	java
[04:10:24] <qsheets_>	brb
[04:10:35] <qsheets>	ok
[04:10:36] <robchurch>	Yeah, but they'll probably get removed during code review.
[04:10:41] <davidmccabe>	The reason we should talk is that I am planning, after we finally get liquidthreads 1.0 out the door, to integrate Campfire-like chat.
[04:11:33] <VoiceOfAll>	robchurch: btw, I wonder if SVN viewvc will support paging
[04:11:35] *qsheets 	is making patch files
[04:11:42] <VoiceOfAll>	I mean these commit logs can get kinda long
[04:11:47] <robchurch>	heh
[04:12:07] <robchurch>	shit, that "We should talk" thing's all ominous
[04:12:08] <davidmccabe>	Ok folks, try this UI on in your mind:
[04:12:11] <qsheets>	and... 2 done 1.6.10 and 1.10.0
[04:12:18] *robchurch 	hopes he never does anything to impress davidmccabe
[04:12:27] <qsheets>	lol
[04:12:46] <davidmccabe>	cool.
[04:12:52] <davidmccabe>	So,
[04:13:02] <davidmccabe>	The "you have a new message on your talk page" message.
[04:13:19] <davidmccabe>	Instead of linking to your talk page, it links to Special:Inbox, which shows all new threads on all talk pages that you are watching.
[04:14:23] <robchurch>	mmm
[04:14:46] <robchurch>	That might later conflict with the new notification thing we're hoping to implement soonish.
[04:15:38] <robchurch>	Where soonish is from one week to six millennia. ;)
[04:15:46] <davidmccabe>	tell me about it
[04:15:56] <davidmccabe>	I might be impressed.
[04:17:07] <robchurch>	I'm not sure I should risk it :P
[04:17:37] <davidmccabe>	I think the benefits are worth the potential costs.
[04:17:53] <robchurch>	http://www.mediawiki.org/wiki/User:Brion_VIBBER/Notification_notes
[04:18:01] <x99>	 But are they worth the actual costs?
[04:18:28] <robchurch>	Now, if we had that, you could maybe fire a notification off saying, "yeah, some discussions you're watching have been changed"
[04:19:02] <robchurch>	Outright changing the new messages thing might not last, if you see what I mean
[04:19:18] <davidmccabe>	well maybe I could implement this.
[04:21:03] *qsheets 	is wondering if he could make code easier
[04:21:28] <qsheets>	ohhh... deleting that line
[04:22:13] *davidmccabe 	goes for a walk.
[04:28:00] <qsheets>	robchurch: should whole new function = +1 sub-version or +1 version
[04:28:12] <robchurch>	what?
[04:29:07] <qsheets>	like at first users couldn't have a preset irc nick in pjirc
[04:29:13] <qsheets>	...and now they can
[04:29:24] <robchurch>	...and?
[04:29:56] <qsheets>	so should the version go from 1.0 to 1.1 or to 2.0
[04:32:34] <qsheets>	...
[04:32:47] <robchurch>	version of what?
[04:32:54] <robchurch>	your product? whatever you want
[04:32:59] <robchurch>	probably 1.0.1 or 1.1.0
[04:34:06] <qsheets>	alright
[04:35:53] <qsheets>	thx
[04:47:29] <CIA-13>	03aaron * r24304 10/trunk/extensions/FlaggedRevs/FlaggedRevs.php: *Use consistent CSS, viciously murder and erase some ampersands. Some cleanup.
[04:51:32] <CIA-13>	03aaron * r24305 10/trunk/extensions/FlaggedRevs/ (FlaggedRevsPage.i18n.php FlaggedRevsPage_body.php): *opps
[05:05:38] <davidmccabe>	qsheets: think of two letters that sound good.
[05:06:06] <davidmccabe>	I recommend NX for "nick exchange".
[05:06:18] <davidmccabe>	Pjirc NX 2.
[05:09:53] <qsheets>	why?
[05:13:06] <davidmccabe>	I am kidding.
[05:26:51] <DFRussia>	I keep having troubles getting some <script> tags in the header of a special page I am making
[05:26:57] <DFRussia>	does anyone have any advise?
[05:27:09] <davidmccabe>	let me look up the function. one sec.
[05:27:21] <DFRussia>	(I want it in the header so I can load a .js file instead of having it in the middle of the php)
[05:27:49] <davidmccabe>	http://pastie.textmate.org/80868
[05:29:33] <DFRussia>	*looks*
[05:30:31] <DFRussia>	I tried something very similar to that before
[05:30:50] <DFRussia>	based on $wgOut->addScript(); but it would not work
[05:31:25] <davidmccabe>	I dunno, it works for me.
[05:32:55] <DFRussia>	ok
[05:33:02] <DFRussia>	I will continue messing with it
[05:33:05] <DFRussia>	thanks for the help
[05:33:09] <davidmccabe>	any time.
[06:26:34] <DFRussia>	http://pastie.textmate.org/80873  <<< does anyone see errors in this JavaScript?
[06:47:28] <CIA-13>	03david * r24306 10/branches/liquidthreads/extensions/LqtExtension.php: correctly display titles in non-main namespaces
[07:12:38] <wikibugs>	03(NEW) Separate the Namespaces in the Categorie View - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=10652 15enhancement; low; MediaWiki: Categories; (michael.frey)
[08:19:10] <arno_>	hi
[08:22:27] <arno_>	I want to define a few javascript functions that can be called for event handling but I don't know were to define them
[08:22:50] <arno_>	I tried Mediawiki:Monobook.js but it doesn't seem to work
[08:23:25] <arno_>	I had no luck trawling through the docs on mediawiki.org
[08:58:31] <_wooz>	lo
[09:01:21] *Evgen2 	have attempt to migrate from 1.6.10 and latin1  unicode's "coding" to 1.10 with utf-8.... with install operation... khe...khe..
[09:35:02] *qsheets 	is Satisfied
[09:35:34] <qsheets>	why?
[09:36:06] <davidmccabe>	Why, precious? Why are we satisfied, precious?
[09:36:38] <qsheets>	...because 1, I'm talking to myself, and , 2, I'm about done coding
[09:39:06] <qsheets>	davidmccabe: I mean you can install the irc now
[09:39:23] <davidmccabe>	oh, I had no interest in your IRC.
[09:39:26] <qsheets>	:-)
[09:39:47] <davidmccabe>	But perhaps I could interest you in AJAX-powered chat with the transcript saved to a comment on a talk page.
[09:39:57] <qsheets>	you just try to annoy, huh?
[09:40:07] <davidmccabe>	I hope I haven't been annoying.
[09:40:39] <davidmccabe>	bye now.
[09:40:49] *qsheets 	hates phpfreechat
[09:41:01] <qsheets>	so...
[09:41:08] *qsheets 	uses PJIRC
[10:40:27] <VoiceOfAll>	_mary_kate_: ping
[10:41:04] <_mary_kate_>	what
[10:41:15] <VoiceOfAll>	can you make me a MW.org sysop?
[10:41:32] <_mary_kate_>	ok
[10:41:37] <_mary_kate_>	whats your username
[10:41:49] <VoiceOfAll>	# Voice of All
[10:41:58] <VoiceOfAll>	opps, just "Voice of All"
[10:42:39] <_mary_kate_>	i made you a sysop and a bureuacrat
[10:45:06] <VoiceOfAll>	ok, thx
[10:45:13] <VoiceOfAll>	now I can delete silly images
[10:48:32] <VoiceOfAll>	_mary_kate_: http://www.mediawiki.org/wiki/Special:Nuke wtf?
[10:48:43] <VoiceOfAll>	that thing gives me the willies
[10:49:06] <_mary_kate_>	brion wrote that to clean up after tards
[10:50:03] <stemd>	last time I knew how mediawiki works was v1.5, now I have a question:
[10:50:05] <stemd>	is there a way to hide some pages from non-registered users (except to delete them, of course ;-)
[10:50:43] <stemd>	(admins can see deleted pages)
[10:51:02] <stemd>	but other can also before they are deleted...
[10:53:19] <VoiceOfAll>	_mary_kate_: btw, how did you feel about the board elections?
[10:54:09] <_mary_kate_>	fine?
[10:55:02] *VoiceOfAll 	is glad that kat won
[10:56:46] <_mary_kate_>	yes
[10:56:50] <_mary_kate_>	i didn't vote for erik.  ;-)
[10:57:48] <VoiceOfAll>	you must be kidding :)
[10:58:01] <VoiceOfAll>	you know he reads any logs/blogs for his name don't you?
[11:00:17] <VoiceOfAll>	kat was the only one that won that I voted for, don't really know Frieda...
[11:00:25] <VoiceOfAll>	still, I'm happy enough
[11:01:24] <VoiceOfAll>	ida_mayhem: :)
[11:03:14] <VoiceOfAll>	_mary_kate_: did you vote for yourself :D ?
[11:03:19] <VoiceOfAll>	be honest
[11:05:48] <_mary_kate_>	i don't remember, but i'd hope i did :)
[11:09:46] <VoiceOfAll>	_mary_kate_: anytime you are about to mention erik, say "on to erik..." and then boot ida_mayhem
[11:09:54] <VoiceOfAll>	it would make a good running joke
[11:37:43] <wikibugs>	03(FIXED) Image upload replacements should create watchlist entry - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=778  +comment (10JSchulz_4587)
[11:39:49] <wikibugs>	03(mod) Namespace and site rename on Norwegian Wikibooks - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=10641  summary; +comment (10herman)
[11:45:24] <wikibugs>	03(NEW) Update for Cantonese language (zh-yue) #47, Update for Old Chinese / Late Time Chinese languages (och/ltc) , Update for Chinese (Taiwan) localisation (zh-tw) - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=10653 15enhancement; normal; MediaWiki: Internationalization; (shinjiman)
[11:45:41] <wikibugs>	03(mod) Update for Cantonese language (zh-yue) #48, Update for Old Chinese / Late Time Chinese languages (och/ltc) , Update for Chinese (Taiwan) localisation (zh-tw) - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=10653  summary (10shinjiman)
[11:57:40] *Evgen2 	just migrate from backwards-compatible utf8 to MySQL utf8
[11:59:39] <robchurch>	Right then, bring on the spoilers!
[12:01:02] <robchurch>	Tut tut, shouldn't be discussing board votes. Top secret political wrangling, that. :)
[12:02:49] <wikibugs>	14(DUP) Separate the Namespaces in the Categorie View - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=10652  +comment (10robchur)
[12:02:50] <wikibugs>	03(mod) Categories need to be structured by namespace - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=450  +comment (10robchur)
[12:02:54] *robchurch 	is well aware he's referring to a 40-minute-old discussion.
[12:02:58] <robchurch>	Perhaps older.
[12:22:40] <lir>	I have uploaded to an ftp directory images/someimage.png and tried to display it with [[Image:images/someimage.png|thumb|center]] but it doesn't work, instead it shows a link and when I click on it, it tries to upload a file...
[12:24:56] <robchurch>	You need to upload files through MediaWiki, since it stores them in a particular fashion, in a particular path, and creates certain metadata entries in the database.
[12:25:13] <robchurch>	You can import images already present on the server using maintenance/importImages.php.
[12:26:50] <lir>	ok importing images that already on the server sounds like a good idea, how do I access that importImages.php page?
[12:27:09] <robchurch>	It's not a page, it's a script to be run from the command line.
[12:27:23] <lir>	I see
[12:28:36] <lir>	hmm, I don't have that but I have other scripts
[12:29:11] <lir>	I have rebuildImages.php 
[12:29:56] <robchurch>	It's possible to use that, although you will have to manually copy all the image files to their correct locations.
[12:30:29] <lir>	ok so what do I do then...?
[12:33:56] <robchurch>	Do you have any existing images uploaded to MediaWiki?
[12:34:37] <lir>	yeah I already uploaded all of the images which I want to show on a page to the images/* directory with extension.png
[12:34:59] <robchurch>	But none were uploaded through MediaWiki? There are no subdirectories in images consisting of cryptic letter combinations?
[12:35:14] <lir>	right, none was uploaded using MediaWiki
[12:35:19] <lir>	that poses a problem?
[12:35:24] <robchurch>	Nope, it makes your life easier.
[12:35:33] <robchurch>	1. Set $wgHashedUploadDirectory = false; in LocalSettings.php
[12:35:35] <lir>	Good :)
[12:35:38] <robchurch>	2. Run maintenance/rebuildImages.php
[12:35:41] <lir>	Just a sec
[12:36:11] <robchurch>	Your images should be found, MediaWiki should tell you off, and then it'll create metadata entries. Then you just use [[Image:Foo.png]] in a page and you're off - full help on image syntax can be found on MediaWiki.org or Wikipedia.
[12:36:32] <Basstard`>	rude mediawiki
[12:37:25] <lir>	I've set that false directive in Localsettings.php and tried running the script, though it exited with error message about problem accessing the mysql database 
[12:37:40] <robchurch>	Create an AdminSettings.php file, see AdminSettings.sample
[12:39:58] <lir>	ok it ran though it reported about 0 of 0 rows updated
[12:40:44] <robchurch>	That's not so good.
[12:41:22] <lir>	hmmm
[12:42:24] <lir>	where is the problem?
[12:42:45] <robchurch>	I'm not quite sure. It could be that the script can't find the images, which might indicate a permissions error.
[12:44:00] <lir>	I set mode 777 on images/
[12:49:14] <robchurch>	Does the script output any errors or other information aside from the zero row count?
[13:00:14] <_DannyB|backup>	is there any way how to acquire message when such mediawiki page doesn't exist?
[13:02:44] <robchurch>	what
[13:02:53] <lir>	robchurch: nope, just: 2007-07-21 12:43:48: Processing image... 2007-07-21 12:43:48: Finished image... 0 of 0 rows updated and same about Processing oldimage
[13:03:05] <robchurch>	d'oh
[13:03:11] <robchurch>	lir: You might need to run it with --missing as an option
[13:03:29] <lir>	let me try that
[13:04:00] <lir>	no that doesn't work either
[13:04:47] *robchurch 	frowns
[13:06:02] <_DannyB|backup>	robchurch: that "what" was on me or on lir?
[13:06:11] <robchurch>	That "what" was for you.
[13:07:27] <robchurch>	lir: The output is more or less identical, apart from the times?
[13:08:12] <lir>	robchurch: yeah
[13:09:01] <robchurch>	You're doing php rebuildImages.php --missing?
[13:09:15] <lir>	yep
[13:09:38] <lir>	wait, I might had a typo
[13:09:50] <lir>	yeah that was one 's'
[13:09:57] <robchurch>	ah
[13:10:13] <lir>	ok I see that it added the files now
[13:10:28] <lir>	thanks
[13:10:36] <robchurch>	whee
[13:10:42] <lir>	so do I just use [[Image:filename.png]] now?
[13:11:26] <lir>	umm, no that doesn't seem to work.
[13:11:50] <lir>	when I do that it shows a link that when I click on it's telling me about MediaWiki error'ed out trying to upload to the directory
[13:12:40] <robchurch>	Is this a public wiki?
[13:17:09] <_DannyB|backup>	robchurch: if the message is taken from mesages.php and not from mediawiki:foo, how can i get that message? i mean i can get page content either from db on toolserver or by action=raw. but how can i get the default messages?
[13:17:37] <robchurch>	There's no external method to get that at this time.
[13:17:56] <robchurch>	Internally, you can use a specific wfMsgXXX() function to bypass the cache and database.
[13:33:46] <_DannyB|backup>	robchurch: how difficult would be solving 6908?
[13:34:54] <robchurch>	Depends on whether our Skin implementations are as crappy as we like to pretend they're not.
[13:35:20] <robchurch>	In other words, whether one can cleanly get the name of the page used for per-skin CSS/JavaScript inclusion out of the Skin object.
[13:52:52] <nstutsman>	who has the patience here to help a "php idiot"?
[14:16:59] <General_Jung>	Hello Guys, I have some question and be very pleased about an helping
[14:17:21] <General_Jung>	I installed some xtension and get this error:
[14:17:22] <General_Jung>	Warning: Cannot add header information - headers already sent by
[14:27:10] <gemidjy>	often while I work on mediawiki site, i get this error: "Sorry! We could not process your edit due to a loss of session data. Please try again. If it still doesn't work, try logging out and logging back in." 
[15:08:14] <Lili>	hi, i want to use my own css file in mediawiki, how to do it
[15:08:55] <robchurch>	Edit the MediaWiki:Common.css page on wiki.
[15:13:43] *habj 	bows to the almighty devs and asks how/where one changes the logo in left upper corner
[15:15:04] <Lili>	thanks, then i want to change the html as well (all links in a table), how to do that ???
[15:15:58] <habj>	we have had the wrong font on the svwiki version of Wikipedia logo...
[15:16:17] <G_SabinoMullane>	habj: http://www.mediawiki.org/wiki/Help:FAQ#How_do_I_change_the_logo.3F
[15:16:34] <habj>	G_SabinoMullane: thanks a lot
[15:17:20] <habj>	G_SabinoMullane: who can manipulate the LocalSettings.php file? Since it is a Wikipedia, is it only the devs?
[15:19:21] <G_SabinoMullane>	Yes, I imagine so: one needs server access.
[15:20:17] <mark>	habj: usually by uploading/replacing [[Image:Logo.png]]
[15:20:18] *habj 	asks a dev, any dev, for help with something that seems to take dev powers
[15:20:30] <habj>	mark: hmmm I could try
[15:21:38] <_DannyB|backup>	habj: to change a logo of wikimedia site you must follow these steps:
[15:21:58] <_DannyB|backup>	upload new logo to that project and protect the page
[15:22:53] <_DannyB|backup>	file a bug saying you want to set the logo for that site named [[Image:<name_you_chosed>]]
[15:22:57] <_DannyB|backup>	that's it
[15:23:08] <_DannyB|backup>	let me show you similar request
[15:23:45] <_DannyB|backup>	bug 8526
[15:24:22] <habj>	perfect! will do
[15:25:55] <_DannyB|backup>	!logo
[15:25:55] <mwbot>	The logo that appears in the top left of each page is determined by the $wgLogo configuration setting in the LocalSettings.php file. To change this you simply need to change the value of $wgLogo to point to the URL of your own logo image. See <http://www.mediawiki.org/wiki/Manual:%24wgLogo> for more information.
[15:26:54] <_DannyB|backup>	there should be similar notice for logo on wmf sites describing what i've just said...
[15:27:43] <Lili>	what file to change if I want to have another structur of my side (the mainpage and all other pages)
[15:29:16] <_DannyB|backup>	what you mean by "another structure"?
[15:30:41] <Lili>	_DannyB|backup: somethink like , put the linkmenu to the right side of page
[15:35:57] <_DannyB|backup>	Lili: edit css files
[15:36:07] <MartinCleaver>	_DannyB|backup: that server of mine (wiki.wikiconsulting.com) is now served by DNS. Can you help with the skin again?
[15:36:22] <Lili>	_DannyB|backup: do you understand my question ??? (maybe this thing is called  html-template )
[15:38:55] <_DannyB|backup>	Lili: if you mean the left sidebar by "linkmenu" then it's possible by editing css
[15:39:14] <_DannyB|backup>	MartinCleaver: remind me, what you need, pls
[15:40:46] <MartinCleaver>	ok, so I copied the monobook directory, the MonoBook.php and MonoBook.deps.php to wikiconsulting, WikiConsulting.php and WikiConsulting.deps.php files
[15:41:21] <lir>	robchurch: yeah it's a public wiki
[15:41:22] <Lili>	_DannyB|backup: i want my whole side in one big table so i can set a backgroundimage , and can have diffrend backgrounds for td's
[15:41:28] <MartinCleaver>	I edited WikiConsulting.php to reflect WikiConsulting:
[15:41:30] <MartinCleaver>	class SkinWikiConsulting extends SkinTemplate {
[15:41:30] <MartinCleaver>	        /** Using monobook. */
[15:41:30] <MartinCleaver>	        function initPage( &$out ) {
[15:41:30] <MartinCleaver>	                SkinTemplate::initPage( $out );
[15:41:30] <MartinCleaver>	                $this->skinname  = 'wikiconsulting';
[15:41:31] <MartinCleaver>	                $this->stylename = 'wikiconsulting';
[15:41:33] <MartinCleaver>	                $this->template  = 'WikiConsultingTemplate';
[15:41:35] <MartinCleaver>	        }
[15:41:37] <MartinCleaver>	}
[15:41:39] <MartinCleaver>	/**
[15:41:41] <MartinCleaver>	 * @todo document
[15:41:43] <MartinCleaver>	 * @addtogroup Skins
[15:41:45] <MartinCleaver>	 */
[15:41:47] <MartinCleaver>	class WikiConsultingTemplate extends QuickTemplate {
[15:41:49] <MartinCleaver>	(sorry, big paste)
[15:41:54] <MartinCleaver>	but the background images etc are not showing
[15:41:54] <^demon>	Please don't paste in the channel.
[15:41:58] <^demon>	Use a pastebin.
[15:42:10] <MartinCleaver>	ok. Will do.
[15:42:12] <Lili>	what is pastebin
[15:42:27] <MartinCleaver>	http://mediawiki.pastey.net
[15:42:34] <^demon>	It's a place you can paste a lot of text, it gives you a URL and you can show people what you pasted.
[15:43:49] <Lili>	in which directory ???  WikiConsulting.php 
[15:44:06] <MartinCleaver>	skins/
[15:44:39] <MartinCleaver>	& skins/wikiconsulting
[15:45:11] <Lili>	sorry i don't have a file named WikiConsulting.php
[15:45:29] <MartinCleaver>	y, Lili, I am making a new skin
[15:45:42] <MartinCleaver>	so I copied the MonoBook structure 
[15:46:09] <MartinCleaver>	but before I change anything of mine I note that it already can't see the bg images
[15:46:27] <Lili>	i copied the dir common and named it test, but i it doesn't work
[15:48:01] <MartinCleaver>	Which instructions did you follow?
[15:49:23] <Lili>	non because i don't know where to change anything, i searched at http://meta.wikimedia.org/ but everything make me more confuss
[15:50:01] <^demon>	Check out www.mediawiki.org
[15:50:20] <MartinCleaver>	Well, I started with http://www.mediawiki.org/wiki/Manual:Skinning
[15:51:38] <MartinCleaver>	Not sure whether I should change:                 wfRunHooks( 'MonoBookTemplateToolboxEnd', array( &$this ) );
[15:51:39] <_DannyB|backup>	Lili: i guess i don't understand your needs. background image can be set to any element
[15:52:45] <Lili>	MartinCleaver:  i don't have the file FooBar.php
[15:52:58] <lir>	I gotta go, bbl
[15:53:03] <Lili>	i will make a picture
[15:53:14] <MartinCleaver>	" Note: The term FooBar will be used as a placeholder for the newly-created skin's name. (Notice the use of the uppercase FooBar and the lowercase foobar in different contexts.)"
[15:54:38] *MartinCleaver 	reads http://meta.wikimedia.org/wiki/Skins
[15:54:40] *MartinCleaver 	changes the Toolbox line
[15:58:17] <MartinCleaver>	still no joy
[15:58:28] <Lili>	MartinCleaver: how can i test this
[16:00:03] <MartinCleaver>	how are you currently testing?
[16:02:09] <Lili>	the mainpage of mediawiki on localhost
[16:03:11] <MartinCleaver>	food is here
[16:03:13] <MartinCleaver>	back in 20
[16:03:31] <Lili>	$this->skinname  = '[foobar]';   .... are this [] necessary  
[16:23:54] <Lili>	.
[16:31:36] <MartinCleaver>	back in a bit (email outage)
[16:32:21] <brlcad>	hello, I'm working on a project where we're trying to use mediawiki for a knowledge database that effectively boils down to a collection of key/value pairs that will likely change over time
[16:33:37] <Duesentrieb>	brlcad: look into semanticmediawiki or the wikidata stuff used by omegawiki
[16:33:40] <brlcad>	is there a way with mediawiki as is, or a plugin, that will effectively provide the functionality of infoboxes, but with consistency enforcement so if someone where to change a base template that was tied to 100's of entries, that all entries would automatically be updated
[16:34:19] <Duesentrieb>	see above
[16:34:21] <^demon>	If you update a base template that is transcluded everywhere, it *will* update them.
[16:34:40] <brlcad>	Duesentrieb: thanks, I'll look into those
[16:34:42] <Duesentrieb>	i assume he's talking about template use, with parameters. i.e. data records
[16:36:24] <brlcad>	yeah, so take an address book entry template for example, and I may want to add a new field a year from now that is "country of origin" or something similar that didn't previously exist -- how to have that show up on all fields automatically -- same does if a field were renamed, or enforced order, etc
[16:39:05] <Duesentrieb>	brlcad: hm? chnaging names or reordered, that shows everywhere immediately. if you add a field, people need to provide a value - you can provide a default, though
[16:39:15] <Duesentrieb>	you don't need anything extra for this type of things
[16:39:45] <Duesentrieb>	you should use named parameters though. and don't change the names, only the labels.
[16:41:35] <brlcad>	what I mean I think is just a difference of visibility, how would a new user visiting a particular address book entry know that there's a new field that needs to be filled in without visiting the template page?
[16:41:57] <brlcad>	and without someone else editing all usages of that template
[16:42:32] <Duesentrieb>	the field would show up empty - or with a message you provide as default.
[16:42:59] <Duesentrieb>	have you actually playedwith templates? it seems to me you are a bit unclear about how they work...
[16:43:01] <brlcad>	ah, then please forgive my ignorance, I'm just starting to get up to speed on what all mediawiki is capable of .. how exactly does that work?
[16:43:30] <brlcad>	the templates I use on my site and that I've used on wikipedia don't behave this way (and perhaps that's just how the templates were defined)
[16:44:01] <Duesentrieb>	brlcad: http://www.mediawiki.org/wiki/Help%3ATemplates#Using_parameters_in_templates
[16:44:10] <brlcad>	e.g. the software template on wikipedia, it doesn't show default empty fields, it just shows those that I fill in, and I had to know about what those fields were
[16:44:50] <Duesentrieb>	that would be because the parameters are explicitely handeled to be optional, i assume
[16:45:04] <Duesentrieb>	to know about *optional* parameters, you'll need to look at the templatepage
[16:45:26] <brlcad>	ah, then perhaps just luck of the draw .. *all* of them that I'd worked with were entirely optional, hence my confusion
[16:45:42] <brlcad>	that's cool then
[16:45:47] <brlcad>	this might just work out of the box
[16:46:25] <lleoliveirabr>	Hi qsheets
[16:46:49] <brlcad>	but then if I understand you correctly, if I needed to add some logic behind those fields, I'd be needing to look more at the semantic mediawiki
[16:47:20] <Duesentrieb>	brlcad: sematic mediawiki and wikidata are different aproaches to maintain actual databases using wiki syntax.
[16:47:47] <Duesentrieb>	"normal" templates just help you create wiki pages - there's no way to actually *query* the data you colelct that way
[16:48:41] <brlcad>	hm, okay
[16:48:46] <MartinCleaver>	Lili: I'm back now
[16:49:10] <Duesentrieb>	brlcad: btw - the help page doesn't seem to mention default values - you can use {{{foo|''please provide the foo parameter''}}}, or something like that
[16:49:24] <Lili>	good ^^ 
[16:50:01] <brlcad>	Duesentrieb: how would one go about adding something like a comparison page, similar to how you can compare edit diffs, but comparing two separate pages?
[16:50:25] <brlcad>	with a specific format, of course, not as low level as the edit diff apge
[16:53:35] <Duesentrieb>	brlcad: write an extension to do it? what would it compare?
[16:57:43] <brlcad>	Duesentrieb: two pages that use the same template, it sould basically show the two side-by-side, hilighting which were different and/or how they were different (e.g. if they were numeric values)
[16:58:24] <brlcad>	I figured that would be some sort of extension/plugin, just making sure it's not something else I could already handle in the front-end without writing the code :)
[16:58:47] <Duesentrieb>	doing this is going to be tricky.
[16:59:08] <Duesentrieb>	the key/value pairs are not stored explicitely anywhere. they are only evaluated for rendering.
[16:59:40] <Duesentrieb>	sounds like wikidata functionality to me - which would probably be a good idea for an address-book type wiki.
[16:59:49] <brlcad>	okay, that's good to know
[16:59:56] <Duesentrieb>	but wikidata is much in flux, and currently mainly aimed at omegawiki (i.e. a dictionary)
[17:00:13] <Duesentrieb>	not even sure if it's maintained separately
[17:00:15] <Duesentrieb>	but i gotta go
[17:00:17] <Duesentrieb>	bbl
[17:00:33] <brlcad>	thanks for your insight
[17:00:37] <brlcad>	much appreciated
[17:06:19] <MartinCleaver>	Damn! It was a caching problem
[17:06:36] *MartinCleaver 	has regained control
[17:07:47] <brlcad>	ah wow, wikidata is exactly what I was looking for
[17:08:27] <wikibugs>	03(NEW) div overflow flags break printable layout - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=10654 major; high; MediaWiki: Page rendering; (cragos)
[17:24:03] <mjung>	hi folks, just a short question, because I was not able to find it: is there a way to include a page or even better just a section of another wiki-page into the current one using mediawiki-markup?
[17:26:23] <^demon>	mjung: Put {{:pagename}}, and it'll include the whole page. Doing a section isn't possible.
[17:28:03] <mjung>	^demon|away: thanks
[17:34:35] <twb>	I'm trying to programmatically download the source of a mediawiki page.  I can't use &action=edit because that requires authentication (login).
[17:34:57] <twb>	Specifically, http://www.haskell.org/haskellwiki/Hpaste.el
[17:36:39] <rainman-sr>	action=raw
[17:37:55] <twb>	Thanks
[17:38:18] <Schroeder2>	Tarik Glenn's retiring?
[17:38:21] <Schroeder2>	wtf?
[17:50:40] <Pathoschild>	Could someone change the default text for the new MediaWiki:Category-empty? "This category contains currently no articles or media" is grammatically incorrect. :)
[17:51:06] <mjung>	I have another (simple) question: a <pre> breakes an ordered list (using *-notation). Is there a way to get around this?
[17:51:44] *Pathoschild 	suggests "This category is currently unused", or "There are no pages or files in this category", preferably with a line or header above to separate it from the category text.
[18:08:34] <VoiceOfAll>	robchurch: http://www.mediawiki.org/wiki/Project:Current_issues
[18:09:11] <VoiceOfAll>	ahh, nevermind, thought that was imported
[18:09:46] <VoiceOfAll>	robchurch: http://www.mediawiki.org/w/index.php?title=Commit_access&diff=next&oldid=111089 :)
[18:31:54] <Avatarms>	hi all
[18:31:55] <Avatarms>	^^
[18:32:17] <Pathoschild>	Hello.
[18:48:15] <Armedblowfish>	Hello, does anyone know what is wrong with Bugzilla?
[18:49:07] <Nikerabbit>	it has too many bugs, perhaps :)
[18:50:05] <Armedblowfish>	'Software error: Can't connect to the database.  Error:  Too many connections'
[18:50:10] *Pathoschild 	goes off to file a bug report on bugzilla: "Can't file bugs on Bugzilla!"
[18:50:52] <robchurch>	mark, domas? Any chance of a quick poke of db7/db8?
[18:51:46] <Armedblowfish>	Hello Pathoschild!
[18:52:16] <Armedblowfish>	Well, does anyone recall from memory if there is a bug report on the topic of entering IPs to block as a list?
[18:52:43] <robchurch>	I recall there being a bug asking for a slightly easier-to-use range block syntax.
[18:52:44] <domas>	whats wrong with db7/db8?
[18:53:04] <robchurch>	er, srv7/8
[18:53:23] <robchurch>	(bugzilla broken? scroll up etc.)
[18:53:26] <Armedblowfish>	robchurch: I was thinking of a situation where not all of the IPs were in the same range.
[18:53:47] <robchurch>	...and you really can't make multiple blocks?
[18:53:48] <domas>	mhm
[18:53:55] <domas>	I didn't touch srv7/8 before at all
[18:54:03] <Armedblowfish>	robchurch: For time-saving.
[18:54:12] <lleoliveirabr>	If i want to anonymous only edit some pages, and can't do nothing more,  is it possible?
[18:54:21] <robchurch>	I don't remember a bug for that.
[18:54:24] <robchurch>	lleoliveirabr: Perhaps.
[18:54:39] <lleoliveirabr>	Still is necessary extensions for this?
[18:55:20] <robchurch>	You might be able to use $wgWhitelistEdit.
[18:55:51] <lleoliveirabr>	Do you think is a great idea create namespaces to add specify different articles
[18:55:54] <lleoliveirabr>	?
[18:55:58] <lleoliveirabr>	Like news articles,
[18:56:09] <Armedblowfish>	robchurch: If I wrote something for it, is there any chance it would get committed? 
[18:56:11] <lleoliveirabr>	Other , companies informations
[18:56:16] <domas>	should work
[18:56:22] <robchurch>	Depends on whether what you wrote was crap or not.
[18:56:59] <lleoliveirabr>	Some people say is not a good idea create namespaces
[18:57:19] <lleoliveirabr>	Is it true?
[18:58:09] <lleoliveirabr>	And is it possible add different namespaces to created articles
[18:58:10] <lleoliveirabr>	?
[18:58:17] <robchurch>	It's probably a good idea to create namespaces if your content is appropriate to partition into namespaces.
[18:58:48] <Pathoschild>	If you do create more namespaces, they should be well-documented to avoid confusion.
[18:58:49] <robchurch>	In other words, if you introduce a different type of "real content".
[18:59:15] <lleoliveirabr>	Example: i want to add information about houses to sell, companies, and news
[18:59:27] <lleoliveirabr>	So probably is a good idea have namespaces
[18:59:29] <TimStarling>	categories are quite decent too
[18:59:55] <lleoliveirabr>	But as i can see is not easy create namespaces
[19:00:07] <robchurch>	Yes; namespaces are probably best suited to anticipated large quantities of different kinds of content.
[19:00:18] <lleoliveirabr>	Oh great
[19:00:24] <robchurch>	Creating custom namespaces is very straightforward; three lines in LocalSettings.php max.
[19:00:57] <robchurch>	(see http://www.mediawiki.org/wiki/Manual:Using_custom_namespaces)
[19:00:58] <lleoliveirabr>	Bur it's it possible to add namespaces to existed articles
[19:01:07] <lleoliveirabr>	Wil read
[19:01:40] <robchurch>	Articles already exist in some namespace. You can move them into the new namespace, or if they're already using a fake prefix, then maintenance/namespaceDupes.php will sort it all out.
[19:05:54] <lleoliveirabr>	And the new pages created? Is it possible to choose which namespace use?
[19:06:50] <lleoliveirabr>	That will be created
[19:31:16] <Monk33>	can I set a magic word like __NOEDITSECTION__ global to all wiki pages?
[19:31:36] <Monk33>	i'm tired of adding it to every page ;)
[19:36:55] <Platonides>	change the default preferences 
[19:37:00] <Armedblowfish>	Where did the includes go?
[19:37:00] <Platonides>	or add the magic css to common.css
[19:37:35] <qsheets>	lleoliveirabr: what did you want?
[19:37:59] <lleoliveirabr>	Are you irc chat extension developer?
[19:38:14] <qsheets>	yeah
[19:38:56] <qsheets>	...why?
[19:39:34] <lleoliveirabr>	Do you know www.wikichat.org
[19:39:36] <lleoliveirabr>	?
[19:41:32] <Monk33>	thx Platonides
[19:42:42] <Armedblowfish>	(If there is no includes directory, then where is blockip.php ?)
[19:43:11] <robchurch>	They're still in includes...?
[19:43:20] <qsheets>	lleoliveirabr: yea, why?
[19:43:42] <lleoliveirabr>	Why you don't work on wikichat
[19:43:43] <lleoliveirabr>	?
[19:43:51] <lleoliveirabr>	Don't you think will be better?
[19:44:57] <qsheets>	idno
[19:45:09] <lleoliveirabr>	What does it idno means?
[19:45:37] <Armedblowfish>	Oh.  Someone could have told me it was in the phase3 directory....
[19:46:16] <qsheets>	lleoliveirabr: it means i don't know
[19:46:34] <lleoliveirabr>	Think about it
[19:46:43] <lleoliveirabr>	qsheets
[19:46:48] <qsheets>	the main reason i created mine was
[19:46:57] <lleoliveirabr>	Would you like to work an a google maps extension
[19:47:01] <lleoliveirabr>	Link:
[19:47:05] <qsheets>	for my robotics team's website
[19:47:05] <lleoliveirabr>	I have some good ideas
[19:47:18] <lleoliveirabr>	Ah ok
[19:47:52] <lleoliveirabr>	I'm not a coder
[19:48:04] <lleoliveirabr>	But the dev said i could you and modify it
[19:48:14] <lleoliveirabr>	http://www.mediawiki.org/wiki/Extension:Ricks_Google_Maps
[19:48:39] <lleoliveirabr>	I have some ideas with semantic forms and semantic mediawiki, to have an advanced filter
[19:48:50] <lleoliveirabr>	Will be very interesting
[19:53:49] *qsheets 	doesn't try to get involved too much
[19:54:45] <lleoliveirabr>	ok
[19:54:54] <qsheets>	lleoliveirabr: it took me ~26 hrs total to get from IRC Chat 1.0 to IRC Chat 1.1
[19:55:16] <lleoliveirabr>	understand
[19:55:18] <qsheets>	...and i'm still coding patches
[19:55:42] <lleoliveirabr>	No problem, i just thought you could interest
[19:55:45] <Monk33>	Platonides do i change the default MagicWord in Local Settings? and what do i need to enter? i found nothing on MW
[19:55:46] *qsheets 	is doing his last main patch right now, for 1.9.3
[19:57:43] <robchurch>	Note that changing the default preference to disable tables of content won't work on browsers not supporting CSS, or where CSS is not generated.
[19:57:54] <robchurch>	er, s/tables of content/section edit links
[19:58:03] <robchurch>	(Actually, it also applies to ToC too)
[20:05:20] <Monk33>	rob you ment me?
[20:06:14] <CIA-13>	03kim * r24307 10/trunk/extensions/Wikidata/OmegaWiki/OmegaWikiAttributes.php: Updated OmegaWikiAttributes to be a (semi-) singleton, will update all other files that were using a global 'till now
[20:15:25] <CIA-13>	03kim * r24308 10/trunk/extensions/Wikidata/OmegaWiki/ViewInformation.php: Added hashCode() method to ViewInformation (needed for OmegaWikiAttributes)
[21:08:35] <lleoliveirabr>	Hi jeluf
[21:11:18] <CIA-13>	03kim * r24309 10/trunk/extensions/Wikidata/OmegaWiki/ (13 files in 2 dirs): 
[21:11:20] <CIA-13>	Removed the $omegaWikiAttributes global entirely, replaced by singleton
[21:11:22] <CIA-13>	OmegaWikiAttributes->getInstance() call.
[21:11:24] <CIA-13>	(Large amounts of search and replace:
[21:11:26] <CIA-13>	please be on the lookout for breakage)
[21:11:28] <CIA-13>	Death to globals!
[21:40:15] <Evgen2>	any body live ?
[21:52:26] <Sparky_>	Can I play "stump the developers" ;)
[22:28:21] <Leefmc>	Im installing a fresh wiki and giving it the DB of my old wiki. Everything runs smooth except the lack of images. What folders should be brought over from the old wiki?
[22:28:37] <Leefmc>	IE, images have to be brought over because they hold the images the DB refers to, but what else?
[22:29:37] <Leefmc>	(and extensions)
[22:32:27] <^demon>	Yes, you need to copy images/extensions/localsettings over.
[22:44:20] <Armedblowfish>	Okay, will someone please tell me,
[22:44:38] <Armedblowfish>	seeing as how you all have not used any of the code the good people of Tor have written for you,
[22:44:43] <Armedblowfish>	why should we bother writing any more?
[22:49:48] <^demon>	Armedblowfish: You've got to realize that the developers don't want to block Tor...it's the community that does.
[22:50:23] <Armedblowfish>	Well, none of our patches got committed.
[22:51:48] <BrokenArrow>	Armedblowfish: have you tried summarizing the issue in a wikitech-l posting? It seems that right now there aren't any devs around
[22:51:57] <^demon>	Armedblowfish: True, the reason being another MakeXXX extension isn't the way to go, a redesign of Userrights was warrented.
[22:52:52] <Armedblowfish>	^demon: There were other patches from years ago.
[22:53:05] *^demon 	shrugs.
[22:53:13] <^demon>	I don't have SVN, I can't help you.
[22:53:31] <Armedblowfish>	There is a post somewhere on the or-talk mailing list mention how discouraging it is.....
[22:53:38] <Armedblowfish>	*mentioning
[22:54:04] *Armedblowfish 	looks for wikitech-l.
[22:55:32] <BrokenArrow>	Armedblowfish: I know the MW team does not communicate well, but people are usually well-intentioned
[22:55:36] <Armedblowfish>	In any case, will someone ask Jayjg to not use 'Tor-using' as a derogatory phrase?
[22:55:58] <lleoliveirabr>	As i can see mediawiki in next version will have a page to add custom namespaces
[22:56:11] <Armedblowfish>	BrokenArrow: I'm not commenting on intentions.  I am saying that this is rather discouraging to us.
[22:56:11] <lleoliveirabr>	But for now i would like how to add this
[22:56:18] <BrokenArrow>	Armedblowfish: we have little control over the whole community, and what the community sees is mainly Tor-using vandals.
[22:56:38] <lleoliveirabr>	And when i create a new article, how choose which namespace use
[22:56:42] <lleoliveirabr>	?
[22:56:44] <Armedblowfish>	We've given you tools to easily block Tor.
[22:56:55] <Armedblowfish>	If you don't use them, what gets through is not our fault.
[22:58:08] <BrokenArrow>	Armedblowfish: I'm not a dev either, just a user and local wiki admin who hangs out a lot here. 
[22:58:26] <Armedblowfish>	Okay.  Can you talk to Jayjg?
[22:59:26] <wikibugs>	03(mod) Use multiple choice listbox for namespace selection - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=10628  +comment (10wikipedia.danny.b)
[23:00:25] <BrokenArrow>	Are his comments on English Wikipedia? I'm not very active there, but I can post a message if you like. (you can do it too, of course)
[23:00:56] <^demon>	BrokenArrow: No he can't, he's been blocked for 2 years because of Tor.
[23:01:51] <wikibugs>	03(mod) div overflow flags break printable layout - 10http://bugzilla.wikimedia.org/show_bug.cgi?id=10654  (10cragos)
[23:01:53] <Armedblowfish>	[[Wikipedia:Administrators'_noticeboard/Incidents#TOR_proxy_editor_reverting_at_House_demolition_in_the_Israeli-Palestinian_conflict]]
[23:01:57] <BrokenArrow>	ca'nt help, sorry. I'm not familiar with en.wiki politics.
[23:02:22] <^demon>	BrokenArrow: It's a very long-running dispute.
[23:02:35] <Armedblowfish>	Just ask Jayjg not to use 'TOR proxy editor' as a derogatory phrase?
[23:03:51] <Armedblowfish>	It's not our fault the editor has 'an inexhaustible supply of TOR proxies'.  We've given Wikipaedia the tools to block Tor effectively.
[23:03:55] <^demon>	Done.
[23:05:15] <Armedblowfish>	Thank you.
[23:06:52] <BrokenArrow>	Thanks demon. English is not my first language, but I must say that in that issue I had not felt there was a derogatory tone (perhaps elsewhere?). The information is factual, and given the current technical issues blocking the page and not the IP was a direct consequence of the IPs being those of TOR exit nodes
[23:08:51] <Pitria>	Hi. Is it allowed to use mediaWiki in a portal with google ads?
[23:10:56] <^demon>	Yes.
[23:11:04] <BrokenArrow>	Pitria: I guess so. see e.g. www.wikia.com 
[23:11:15] <Pitria>	thank you
[23:11:21] <TimStarling>	I'm getting deadlocks due to the ajax upload dest check
[23:12:41] <TimStarling>	could be windows specific, I'm not sure
[23:13:42] <BrokenArrow>	can we test it somehow?
[23:13:57] <TimStarling>	the two locks involved appear to be the session and the thumbnail file
[23:14:19] <TimStarling>	you just upload a file which already exists, and click the submit button while the ajax request is loading
[23:15:02] <BrokenArrow>	is this live on the site or in a svn branch?
[23:15:38] <TimStarling>	I saw it while I was writing the feature, but I ignored it
[23:15:44] <TimStarling>	so it's probably on the live site too
[23:15:48] <Armedblowfish>	TimStarling: Re [[Wikipedia_talk:Advice_to_users_using_Tor_to_bypass_the_Great_Firewall#Hysterics_from_SJMurdoch]]
[23:15:52] <TimStarling>	but we don't lock sessions there
[23:16:00] <Armedblowfish>	You realise Tor uses multiple layers of encryption?
[23:16:02] <TimStarling>	or thumbnail files for that matter
[23:16:13] <TimStarling>	tor uses one layer of encryption: SSL
[23:16:29] <TimStarling>	doesn't make it secure
[23:18:17] <TimStarling>	I'm thinking about trying to defer session start, it wouldn't be a complete fix though
[23:18:28] <TimStarling>	but I'm not sure how to debug this on windows, not many tools
[23:18:41] <Armedblowfish>	TimStarling: A layer of encryption for each node of the chain.  Technically not SSL, but close.
[23:18:44] <Armedblowfish>	And no, it's not perfectly secure, but it's still fairly good.
[23:18:49] <Armedblowfish>	Assuming you aren't using third-party plugins such as Javascript that can phone home.
[23:19:32] <TimStarling>	it uses HTTPS, that's pretty damn close to SSL
[23:19:57] <Armedblowfish>	It still has a higher blocking resistance than secure.wikimedia.org....
[23:19:58] <Avatarms>	notte
[23:20:10] <TimStarling>	how so?
[23:20:25] <TimStarling>	secure.wikimedia.org isn't blocked by port, is it?
[23:20:45] <TimStarling>	it's just as easy to block tor as it is to block secure.wikimedia.org
[23:20:52] <TimStarling>	they just haven't bothered yet
[23:21:26] <TimStarling>	it's just as easy to block tor as it is to block secure.wikimedia.org
[23:21:29] <TimStarling>	they just haven't bothered yet
[23:21:59] <TimStarling>	open proxies are better, because there are more entry points available
[23:22:07] <Armedblowfish>	TimStarling: Tor is a socks proxy, not an https proxy.
[23:22:13] <Armedblowfish>	Not quite as easy, which is probably why they haven't bothered.
[23:22:21] <TimStarling>	Tor is a socks proxy locally, but internode communication is by HTTPS
[23:22:37] <TimStarling>	we block tor on wikipedia easily enough, don't we?
[23:22:53] <Armedblowfish>	*almost* SSL... there are some differences if you look closely.
[23:23:04] <TimStarling>	it's trivial for an organisation the size of the GFW
[23:23:06] <Armedblowfish>	TimStarling: Well, you would if you used the code we gave you.
[23:23:17] <TimStarling>	would what?
[23:23:25] <Armedblowfish>	TimStarling: Tor does have a lot of entry points available.
[23:23:40] <TimStarling>	there are tens of thousands of open proxies
[23:23:46] <TimStarling>	no comprehensive list exists
[23:23:50] <Armedblowfish>	TimStarling: If you all used the code the good people of Tor gave you, you would block Tor more effectively.
[23:23:52] <TimStarling>	you can't say that about tor entry nodes can you?
[23:24:01] <Armedblowfish>	TimStarling: Circumventor is fairly good.
[23:24:37] <Armedblowfish>	TimStarling: No, Tor does not have tens of thousands of entry points, but for one proxy network, it has a lot.
[23:25:13] <TimStarling>	it doesn't matter anyway
[23:25:15] <Armedblowfish>	(Circumventor makes discovery of proxies difficult, so it is pretty good at getting around firewalls, if you can but find the proxies.)
[23:26:23] <TimStarling>	if you're in china, there are various circumvention methods you can use to get access to the bulk of the internet
[23:26:33] <Armedblowfish>	TimStarling: Considering Tor users can pick their exit points, all they have to do to get around Wikipaedia's Tor blocks is find the ones you missed and tell Tor to use them.
[23:26:34] <TimStarling>	but most users don't know about them or don't bother
[23:27:24] <Armedblowfish>	I know. http://cvs.seul.org/viewcvs/viewcvs.cgi/*checkout*/tor/trunk/doc/design-paper/blocking.html?rev=10168&root=Tor
[23:27:54] <BrokenArrow>	TimStarling: does the submit action have a shortcut? the ajax query loads too fast to get the lock on my system...
[23:28:51] <TimStarling>	in firefox, you can click submit during the typing delay, which i think is 500ms
[23:29:10] <TimStarling>	then the form starts submitting, and then the ajax request is sent after the post
[23:29:24] <TimStarling>	that seems to trigger the deadlock pretty reliably
[23:30:10] <TimStarling>	but like I say, it won't happen if you have sessions in memcached, or if you're using NFS
[23:30:37] <Armedblowfish>	TimStarling: Also, if you *were* a dissident, Tor may not be perfect, but it should still provide better security than a single proxy.  Tor provides remote DNS, so unless the eavesdroppers break the encryption or otherwise compromise the Tor network, they won't know where the user is going.
[23:31:13] <TimStarling>	if you were a dissident, you wouldn't have to cross the firewall
[23:31:38] <TimStarling>	you can just set up your dissident websites locally
[23:31:53] <TimStarling>	it's the foreign content that the authorities want to keep out
[23:33:29] <Armedblowfish>	... But it's safer to set them up as hidden services.  : )
[23:34:08] <BrokenArrow>	TimStarling: I have a vanilla install from svn trunk right now, local fs
[23:34:33] <TimStarling>	there's no tor that operates exclusively inside china is there?
[23:34:55] <TimStarling>	so if you wanted to shut down a hidden service, all you'd have to do is block all tor traffic as it traverses the firewall
[23:35:00] <TimStarling>	most circuits would intersect it
[23:35:29] <Armedblowfish>	You'd have to find the hidden service first.
[23:35:43] <TimStarling>	why?
[23:35:56] <Armedblowfish>	Well, it's hidden, isn't it?
[23:36:05] <TimStarling>	it's hidden within the tor network
[23:36:12] <TimStarling>	so if you shut down the whole tor network, it stops working
[23:36:45] <Armedblowfish>	Yes, but then, they haven't bothered blocking Tor.
[23:37:30] <Armedblowfish>	(Also, hidden services often stop working any way... the Tor developers haven't figured out why.)
[23:38:43] <Basstard`>	the east are like "don't do this or you'll get caught." the west are like "do this and you'll get caught."
[23:38:55] <TimStarling>	they haven't bothered blockign tor because there haven't been any hidden services worth blocking
[23:39:03] <TimStarling>	it's naive to assume that tor will solve all of china's problems
[23:39:41] <Armedblowfish>	No, it won't, not as currently set up, but so far it's been doing well.
[23:39:58] <TimStarling>	what, it's been doing well solving all of china's problems?
[23:40:42] <Armedblowfish>	No, just the problems of Chinese Tor users.
[23:40:51] <Armedblowfish>	... on the internet....
[23:41:39] <TimStarling>	right, well I wasn't talking about the problem of entry node discovery
[23:41:43] <mark>	you two are great entertainment, along with some popcorn
[23:41:49] <TimStarling>	I was talking about democracy and freedom of speech and what not
[23:42:21] <TimStarling>	health care, poverty
[23:42:29] <rainman_sr>	people tell me it's getting better.. (as for freedom, etc.. )
[23:42:30] <Armedblowfish>	I'm just talking about the problem of entry node discovery.
[23:42:41] <Armedblowfish>	The rest is really out of Tor's arena.
[23:42:41] <TimStarling>	it most certainly is getting better, but not because of tor
[23:43:25] <Armedblowfish>	Infranet sounds like a good idea, but of course it'd be near impossible to find....
[23:44:30] <Basstard`>	one of the biggest mistakes a nation can do, is to force it's culture and values on to another nation.
[23:45:57] <Armedblowfish>	(Infranet would be a steganography proxy.)
[23:47:25] <sacolcor>	Does anyone have a pointer to information on modding a skin to fix the position of the header and sidebar?
[23:48:03] <sacolcor>	(fix as in CSS position: fixed, not fix as in repair)
[23:48:10] <BrokenArrow>	TimStarling: I can't get that thing to deadlock on win
[23:48:28] <TimStarling>	ok
[23:51:23] <TimStarling>	BrokenArrow: what version of apache?
[23:52:45] <BrokenArrow>	TimStarling: 2.0.58 - let me do another thing...
[23:53:28] <_DannyB|backup>	hello, is #wikimedia-tech dead during holidays?