[00:28:25] <robla>	brion: do you have time to look at https://bugzilla.wikimedia.org/show_bug.cgi?id=34539 ?
[00:28:40] <robla>	AaronSchulz started looking at it, but some swift stuff took priority
[00:29:04] *AaronSchulz 	isn't looking at swift atm
[00:29:35] <brion>	hm
[00:29:52] <robla>	brion: maybe Aaron's got it
[00:30:06] <brion>	'global oversight' must be new in the last couple years, don't remember that feature. ic an take a peek through if you want though :D
[00:35:56] <hexmode>	robla: I've been meaning to ask about the 1.19 schedule, but I guess it can wait till TL;DR in the morning
[01:26:52] <Krinkle>	Nikerabbit: still there?
[01:28:19] <Krinkle>	Reedy: You there?
[02:05:33] <hexmode>	Krinkle: who would you recommend for mysql tests?  I'm thinking chad...
[02:05:47] <Krinkle>	Ask him :P
[02:05:49] <Krinkle>	^demon:
[02:05:51] <Krinkle>	I have no clue
[02:05:55] <Krinkle>	What kind of tests?
[02:05:59] <hexmode>	huh
[02:06:04] <hexmode>	I didn't see him
[02:06:07] *hexmode 	is blind
[02:06:11] <^demon>	Why do I feel like I'm getting volunteered for something?
[02:06:19] <Krinkle>	hexmode: don't worry, he joined while you were typing
[02:06:24] <hexmode>	https://bugzilla.wikimedia.org/34762
[02:06:27] <Krinkle>	like 1 sec before
[02:06:39] <hexmode>	^demon: could you look at that tiny patch
[02:07:01] <hexmode>	looks like it would be a good one for 1.19 tarball
[02:07:29] <hexmode>	Krinkle: yeah I got annoyed with join/part a long time ago and hid them
[02:07:37] <^demon>	I'm not entirely sure why you'd call close() twice anyway ;-)
[02:07:37] <Krinkle>	aha
[02:08:11] <^demon>	$dbw->close(); // close the connection
[02:08:16] <hexmode>	^demon: neither am I... so if you do should you get the warning?
[02:08:16] <^demon>	$dbw->close(); // making sure
[02:08:25] <^demon>	Well yeah
[02:08:33] <^demon>	I'm more curious why close() is getting called twice.
[02:08:45] <^demon>	Probably just a hypothetical, or some awful extension.
[02:09:16] <hexmode>	yeah, I'll ask on the bug
[02:09:28] <hexmode>	unless you've already typed something up?
[02:10:38] <^demon>	I'm about to commit.
[02:10:43] <Krinkle>	^demon: Well, it shouldn't give a low-level PHP warning for msyql_close
[02:10:51] <^demon>	I don't think it's worth backporting, unless he can point to something silly that's already doing this.
[02:10:58] <Krinkle>	no matter what, and the documentation says it should be cleared
[02:11:02] <^demon>	If it's just a hypothetical, it can wait.
[02:11:28] <hexmode>	k, so the question was worth it!
[02:11:33] <Krinkle>	it's a minor case, shouldn't be in production code. (or actually hardly can be since it emits a PHP warning)
[02:11:34] <^demon>	whoops, no git commit.
[02:13:37] <^demon>	Actually, all the database subclasses suffer from this.
[02:15:05] <Krinkle>	^demon: https://wikitech.wikimedia.org/view/Formey , can you confirm that 'will be' is 'is' now ?
[02:15:14] <Krinkle>	"which will be the new server for our public subversion repository, replacing mayflower"
[02:15:54] <^demon>	Oh yeah, mayflower was decommissioned like a year ago
[02:16:02] <Krinkle>	ok, thx
[02:16:19] <^demon>	And actually we'll be moving gerrit off formey to some new box.
[02:17:39] <Krinkle>	ok
[02:19:00] <Krinkle>	^demon: I'm trying to confirm my email for my wikitech account but can't. Something is probably misconfigured in the smtp ?
[02:19:27] <^demon>	No clue. I don't have access to that.
[02:20:16] <Krinkle>	k
[02:20:24] <Krinkle>	who does?
[02:27:51] <^demon>	someone from ops.
[02:28:03] <^demon>	wikitech's hosted offsite, so probably a root user.
[02:28:05] <^demon>	Roan might know
[02:28:21] <RoanKattouw>	Someone who's root on wikitech
[02:28:23] <RoanKattouw>	i.e. not me
[02:28:25] <robla>	Krinkle: try asking on #wikimedia-operations
[02:28:33] <TimStarling>	I'm probably root there
[02:28:54] <robla>	...or just wait for Tim here  :)
[02:30:04] <TimStarling>	2012-02-28 02:19:08 1S2Ceh-0003AL-Uw ** ttijfhof@wikimedia.org R=dnslookup T=remote_smtp: SMTP error from remote mail server after RCPT TO:<ttijfhof@wikimedia.org>: host mchenry.wikimedia.org [208.80.152.186]: 550 Address ttijfhof@wikimedia.org does not exist
[02:30:47] <RoanKattouw>	TimStarling: tijhof not tijfhof
[02:31:00] <TimStarling>	tell Krinkle not me
[02:31:09] <RoanKattouw>	lol
[02:31:20] <TimStarling>	you'd think he'd be able to spell his own name
[02:31:25] <RoanKattouw>	You'd think so yeah
[02:35:37] <AaronSchulz>	TimStarling: are you the same guy that thinks people should be able to count to 12?
[02:35:49] <Krinkle>	TimStarling: Thanks!
[02:36:50] <Krinkle>	TimStarling: mail fixed, account confirmed
[02:36:58] <TimStarling>	some day we're going to have bounce processing in MediaWiki, right?
[02:37:12] <Krinkle>	:)
[02:38:23] <TimStarling>	RoanKattouw: thanks for the bug fix
[02:38:39] <TimStarling>	maybe the lack of dollar signs fooled some part of my brain into thinking I was writing C++
[02:38:55] <RoanKattouw>	heh
[02:39:01] <RoanKattouw>	I figured that's where that came from
[02:39:10] <RoanKattouw>	The way global variables work in JS is really kind of evil
[02:40:07] <TimStarling>	in C++ you have a hugely complex name resolution process,
[02:40:35] <TimStarling>	if it was an interpreted language it'd probably be as slow as JS
[02:40:57] <TimStarling>	it's just the fact that it's done at compile time that saves it
[02:41:21] <RoanKattouw>	Name resolution is probably equally bad in both languages
[02:41:36] <RoanKattouw>	JS doesn't have as much OOP-related complexity, but it has closure scope-related complexity instead
[02:42:18] <TimStarling>	closure scope and OOP are kind of mixed
[02:42:27] <TimStarling>	actually I bet that was where that bug came from
[02:42:29] <RoanKattouw>	In practice they often are yes
[02:42:44] <TimStarling>	I was assuming isReady was a private variable from the enclosing scope
[02:42:47] <RoanKattouw>	If that class were written with a different pattern, what you did would have been right
[02:42:49] <RoanKattouw>	Exactly
[02:43:51] <TimStarling>	aren't you glad PHP has a "class" keyword and doesn't require you to construct OOM from primitives?
[02:43:58] <TimStarling>	OOP
[02:44:08] <TimStarling>	finger memory
[02:44:42] <RoanKattouw>	Yeah
[02:44:47] <RoanKattouw>	I have mixed feelings about JavaScript
[02:44:55] <RoanKattouw>	It's both the most awesome and most terrible language ever
[02:45:08] <TimStarling>	closures are super awesome
[02:45:17] <TimStarling>	which is why they've been imported into PHP I guess
[02:45:38] <RoanKattouw>	(BTW the reason a global variable reference is slightly slower is that it's really doing window.jQuery, so there's a hashtable key lookup involved)
[02:46:03] <RoanKattouw>	Or window['jQuery'] is closer to how it's actually implemented
[02:46:28] <TimStarling>	Krinkle tells me that a global variable reference has to check the local symbol table of every enclosing lexical scope individually
[02:46:45] <TimStarling>	so it can be 4 or 5 hashtable lookups if you have a lot of enclosing scopes
[02:47:04] <RoanKattouw>	Oh, yeah, there's that too
[02:47:09] <RoanKattouw>	You have to make sure it's not in some other scope
[02:48:02] <Krinkle>	yeah, and to make it more complicated: there is no global scope in JavaScript
[02:48:28] <Krinkle>	browsers create a variable var window = new DOMWindow(); in their initial scope
[02:48:47] <AaronSchulz>	TimStarling: where does wmerrors log it's stuff?
[02:48:48] <Krinkle>	and every <script> is executed as with ( window ) { /* script */ }
[02:48:59] <TimStarling>	fenari:/var/log/mw/fatal.log
[02:49:07] <TimStarling>	[28-Feb-2012 02:32:16] Fatal error: Call to a member function getNamespace() on a non-object at /usr/local/apache/common-local/php-1.19/includes/specials/SpecialMovepage.php on line 59
[02:49:08] <RoanKattouw>	Right, that file
[02:49:09] <Krinkle>	so that all properties of window are visible as local variables inside <script> tags
[02:49:11] <TimStarling>	I'm going to look at that
[02:49:16] <RoanKattouw>	I should get around to logrotating it
[02:49:34] <TimStarling>	there's an RT ticket
[02:50:17] <RoanKattouw>	I know, and I have a partial fix
[02:50:18] <AaronSchulz>	PHP Fatal error:  Call to a member function getTimestamp() on a non-object in /usr/local/apache/common-local/php-1.19/extensions/Collection/Collection.body.php on line 565
[02:50:25] <AaronSchulz>	any one ever look at those?
[02:50:32] <TimStarling>	not yet
[02:51:37] <^demon>	Goodnight everyone.
[02:51:49] <AaronSchulz>	au revoire
[02:55:20] <AaronSchulz>	$revision = Revision::newFromTitle( $title, $oldid );
[02:55:33] <AaronSchulz>	hmm, I guess the revs didn't exist
[02:55:45] <AaronSchulz>	'timestamp' => wfTimestamp( TS_UNIX, $revision->getTimestamp() ),
[02:58:54] <Krinkle>	r111983 | TimStarling
[02:58:57] <Krinkle>	currentFocused = $( '#wpTextbox1' ); is still outside the dom-ready hook
[02:59:00] <Krinkle>	!r r111983 | TimStarling
[02:59:00] --elephant--	TimStarling: http://www.mediawiki.org/wiki/Special:Code/MediaWiki/r111983
[02:59:01] <mw-bot>	TimStarling: https://www.mediawiki.org/wiki/Special:Code/MediaWiki/r111983
[02:59:08] <TimStarling>	duplicate bot
[02:59:19] <Krinkle>	RoanKattouw: elephant seems to be pretty hard-headed
[02:59:25] <Krinkle>	still coming back
[02:59:36] <RoanKattouw>	wf
[02:59:49] <TimStarling>	I have a firefox keyword "r", I just paste in the revision number to the location bar and put a space betwen the "r" and the number and it gives me the CR page
[02:59:56] <TimStarling>	so you don't need to give me the URL
[03:00:04] <Krinkle>	TimStarling: yeah, me too actually :)
[03:00:20] <Krinkle>	TimStarling: perhaps that line can be removed ? it's set on-focus of any textarea from the dom-ready hook