[00:03:18] RoanKattouw: edit.js fail - look at line 60 [00:03:35] or, 55 [00:03:41] well, mostly 60 [00:03:53] can't do selection on hidden textarea [00:04:21] the textboxes[0].selectionStart is particularly bad [00:04:55] Isn't that intended to detect a hidden textarea? [00:05:30] no, it's just looking for textarea elements, visible or not [00:05:49] Yes but the textboxes[0].selectionStart === null check [00:05:59] I think that'd trigger for invisible textareas in Firefox [00:06:22] *RoanKattouw notices how copying wikitext from one Wikieditor to another doubles newlines [00:06:29] Wasn't that one of our bugs? [00:16:52] nimish_g and I is close to a solution [00:16:54] RoanKattouw: I'm working on that now [00:16:56] s/is/are [00:17:13] Cool [00:17:27] I have now confirmed that guillom 's bug is in the delayedChange code flow somewhere [00:17:55] RoanKattouw: any claims otherwise have yet to be proven! [00:29:08] Hm I can't reproduce it consistently [00:29:15] Seems to be linked to the enter key inserting two lines [00:33:01] TrevorParscal: How close are you guys to fixing that bug? I'm gonna go to sleep soon and there's two ready-made fixes (Adam's tab fix and my

eating fix) that we should probably deploy asap [00:35:37] hi, i'm back [00:35:46] and will be around for 25 minutes [00:35:57] RoanKattouw: so, the strip down on paste is working [00:36:07] but now when you copy-paste within the wikieditor [00:36:20] you get each arbitrary new line as a

[00:36:30] so, it's even worse looking [00:36:58] i may have misspoke [00:38:03] TrevorParscal: are there two more bug fixes which are ready to be deployed? [00:38:27] if

eating bug is not a prime time for deployment [00:38:38] let's deploy tabbing fix [00:38:52] RoanKattouw: is adam's fix staged? [00:38:59] Nope neither is staged [00:39:09] nkomura: How is that not prime time for deployment? It's a very serious issue [00:39:15] Much more serious than the tabbing issue [00:39:39] *RoanKattouw stages both fixes on deployment prototype [00:39:42] i'm not saying it is less important [00:39:54] if you all not on the same page on the solution, [00:40:02] it is not prime time for deployment [00:40:17] How are we not on the same page? Our discussions are about something else? [00:40:20] -? [00:42:05] it wasn't clear from the tread [00:42:23] Right [00:42:32] TrevorParscal: So how do you feel about the

/  eating fix? [00:42:47] i'll go over to talk to him [00:44:00] can I have revision numbers? [00:44:03] then you can go to bed [00:44:04] :) [00:45:08] I've staged trunk on deployment [00:45:19] Basically I wanna merge to trunk state [00:46:11] That's everything from r62000 upwards [00:46:29] I'll grab a list of the important ones though [00:46:54] r62139, r62141, r62145 for Adam's tab fix [00:47:11] r62143, r62146 for my

/  eating fix [00:49:46] RoanKattouw: will you clarify the scale of this deployment? [00:50:04] It's five revs in total, fixing the tab issue and the

eating issue [00:50:13] Revs are listed just above your question [00:50:16] if we are merging trunk, we do need to run the full suite of tests [00:50:23] Yeah so let's not do that yet [00:51:03] Let's just confine this deployment to those five revs [00:51:18] i feel better with that [00:54:58] RoanKattouw: it's almost 2am for you [00:55:01] Yeah [00:55:14] TrevorParscal: Can you poke me when you've reviewed those revs so I can deploy them and go to sleep? [00:55:26] oh [00:55:26] yes [00:55:51] are the fix staged on deployment? [00:57:08] Yes [00:58:15] &esc; [00:58:30] is that for sure not an existing entity? [00:58:35] Yeah I had to pick *something* [00:58:46] totally, just checking [00:59:01] cause if it was an actual entity that did something real... [00:59:05] Doesn't even matter because it only gets fed into .text() not .html() [00:59:10] So it comes out as &esc; [00:59:11] ah! [00:59:12] well [00:59:23] unless that entity was used in the article [00:59:42] Which is why we escape existing occurrences of &esc; to &esc;esc; first [00:59:51] right [00:59:52] ok [00:59:56] yes, that's sane [01:00:06] http://www.mediawiki.org/wiki/Special:Code/MediaWiki/r62141 [01:00:10] you said it was adams [01:00:14] but it's clearly yours [01:00:25] That's a typo [01:00:29] was there another rev of adams you meant to type but was off by one? [01:00:31] yes.. i suspected [01:00:34] tabbing is working nicely on FF3/Ubuntu [01:00:42] Yeah it's the next one, 142 [01:02:31] i think we should RoanKattouw sleep [01:02:55] ok [01:03:01] your stuff looks good [01:03:08] we can revisit tomorrow or ask werdna for help deployment [01:03:17] :) [01:03:25] go to bed RoanKattouw ! [01:03:27] eek! :) [01:03:28] thank you SOOO MUCH [01:03:36] I've still got 5 hours of work time, feel free to throw stuff my way [01:03:48] thank you RoanKattouw! [01:03:50] it's all part of making you guys feel like you owe me some liquidthreads love in april :) [01:04:16] werdna: So remember when pushing JS changes for us you have to bump the $wgStyleVersion appendix in CommonSettings.php or bits will serve stale JS from pmtpa [01:04:29] RoanKattouw: I shall [01:05:10] i'll be in the meeting TrevorParscal, so please make a call for the deployment [01:05:29] i didn't test the

eating bug on IE as I don't have my other computer with me [01:05:38] but tabbing looked good to me on deployment [01:05:48] Well, I think the only thing worse than running knowingly bad code on the site is pushing an update at 5pm [01:06:03] so I personally would say let's deploy tomorow morning [01:06:09] then we can react all day [01:06:41] i'm getting lots of frustration about the tabbing bug [01:06:59] *RoanKattouw reminds people that the other bug actually messes up pages [01:07:15] Also, these fixes were ready around 3:30 pm already, they should've been pushed earlier [01:07:16] you are talking about the IE bug? [01:07:29] The

eating bug happens in all browsers [01:07:41] It means that if you edit an article that contains

s, they will disappear [01:07:47] *werdna tries to resist making "om nom nom" noises. [01:08:08] okay, let's deploy tomorrow then [01:08:21] werdna: you are rescued [01:08:30] got to run to a meeting [01:08:35] it's fine, I don't mind helping out [01:09:58] *guillom eats werdna [01:12:58] I swear every photo of Brion is the same [01:13:00] http://commons.wikimedia.org/wiki/File:Wikimedia_SF_11th_meet-up_2010-02-06_042.jpg [12:36:21] yike [12:36:23] s [12:36:31] found a bug: http://en.wikipedia.org/w/index.php?title=MediaWiki:Edittools&curid=5536741&diff=342909563&oldid=340526986 [12:37:32] thedj: https://bugzilla.wikimedia.org/show_bug.cgi?id=22435 [12:37:45] RoanKattouw: thx [12:37:48] I fixed it on trunk, we'll deploy it in the morning (PT) [12:41:16] hi RoanKattouw [12:41:19] you're awake I see [12:45:32] Yeah [15:27:08] RoanKattouw: I like your blog post. [15:27:22] why is 22311 a partial fix? how can I complete it? [15:27:49] According to Calcey it doesn't work in Safari 4 or Chrome 5: they insert a tab [15:28:07] I don't have either browser so I can't confirm that [15:28:25] And I'm not sure the code they tested was current; updating prototype to trunk state now [15:29:00] yeah i'm replying to that comment [15:29:34] they claimed it doesnt work in safari 5, but safari is still at version 4 :) [15:29:40] probably just a typo, but still :) [15:30:09] hahaha [15:30:19] it works for me locally in both safari and chrome so i'd assume my fix didnt make it onto prototype before they tested [15:32:10] Could be [15:32:13] Should be live on prototype now [15:33:08] adam_miller: Feel free to follow up on the bug telling them to hit prototype again [15:36:10] adam_miller: Also, could you try to reproduce the copypaste bugs mentioned in my draft? Some of them may have been fixed with Nimish's commit last night, but I was asleep at the time [15:36:21] sure [15:36:47] I'd do it but I need a break from work and figured I'd take it before office hours PT [16:03:19] RoanKattouw_away: seems like all of the copy + paste errors are fixed save for the line breaks in safari [16:03:55] I can't get chrome or firefox to screw up line breaks, but safari does it with text i pasted in from an email [16:04:45] *thedj wonders if there is no hidden feature to "convert to non-formatted text" in webkit... [16:20:43] adam_miller: Awesome! [16:21:16] adam_miller: What's the line breaks in Safari thing exactly? Does it mess up on paste or on save? [16:21:27] on paste [16:21:46] after the text is reformatted, there are no line breaks [16:23:47] Ah [16:24:38] also noticed if i paste text from TextMate, it retains the formatting, but from an email and it does not [16:25:07] It's not supposed to retain formatting ever when pasting into the iframe [16:25:08] both strings seem to be indicating line breaks with \n characters when I paste them into the command line though [16:25:20] sorry, not formatting, line breaks [16:26:16] \t characters seem to be retained as well...not sure if thats desirable [16:26:23] i'm thinking no [16:27:05] \t has been preserved in the past though [16:27:14] Yes, \t should be preserved [16:27:21] Use case: pasting source code [16:27:21] ...for 

 blocks
[16:27:39] 	onea related note: \t at the start of a line should trigger pre-format, like space does :)
[16:28:06] 	Daniel_WMDE: You know who you have to bribe for that, right? Hint: he lives in Australia
[16:28:25] *RoanKattouw 	waves at nkomura
[16:28:44] 	Andrew? :P
[16:33:22] *nkomura 	waves back to RoanKattouw
[16:33:33] 	yay on the deployed fix RoanKattouw
[16:33:37] 	great job!
[16:35:39] 	:)
[16:35:55] 	Adam tells me Nimish's fix from last night fixes most of the pasting issues, except on Safari
[16:36:41] 	So I was thinking we could have Trevor and Nimish poke at that when they get in
[16:36:59] 	Then when it works for Safari as well we can push that fix
[16:37:15] 	it is great hear that the solution works for most of browsers
[16:37:22] 	what does safari do differently?
[16:37:36] 	nkomura: I'd also like you to look at the draft blog post I sent you; I wanna post that as soon as you approve it
[16:37:51] 	thanks for preparing that
[16:38:01] 	i wanted to put together something last night
[16:38:18] 	Also, Calcey reported that Adam's tabbing fix was broken for Chrome and Safari but that was fixed in the meantime; according to Adam it works now
[16:38:28] 	So I wanna deploy that in a few hours
[16:38:30] 	but i had to preserve my 4.5 hrs of sleep to be functional today
[16:38:42] 	k
[16:38:42] 	Oh yeah sure
[16:39:13] 	have you staged the fixed and confirmed with all browsers?
[16:39:22] 	I staged it, Adam confirmed it
[16:39:32] 	code review?
[16:39:42] 	Adam wrote the code, I reviewed it
[16:39:48] 	good morning adam_miller
[16:39:59] 	good morning nkomura
[16:40:13] 	(BTW, I won't be available much this afternoon; I'll be leaving around 1 PM PT, off to bed early for my driving test tomorrow)
[16:40:26] 	glad to hear that tabbing is working adam_miller
[16:40:49] 	So let's hope it doesn't snow tonight :) it was rescheduled twice before due to snow
[16:41:25] 	RoanKattouw: yeah we should stop keeping you up so late
[16:41:55] 	severe storm is coming to the east coast
[16:42:02] 	is it the case in europe too?
[16:42:13] 	No, no storm
[16:42:38] 	adam_miller: have you confirmed that tabbing works for supported browsers?
[16:42:49] 	But it's been freezing again, it'll get down to -8 C (~17F I think) tonight and they think it'll maybe snow a little bit
[16:43:06] 	RoanKattouw: that sounds brutal
[16:43:15] 	nkomura: it does, i checked safari and chrome this morning as those were the two reported as still being broken
[16:43:21] 	i do not want to take my driver's test on the icy surface
[16:43:29] 	adam_miller: great
[16:43:32] 	Oh they won't let me
[16:43:49] 	RoanKattouw: let's deploy the fix for tabbing
[16:43:58] 	many people are waiting for that to be fixed
[16:44:04] 	OK rolling
[16:44:42] 	btw nkomura we got our first proposal from a UX firm
[16:49:51] 	adam_miller:  var $tabindexList = $j( '[tabindex]:visible' ).sort( function( a, b ) { return a.tabIndex > b.tabIndex ? 1 : -1; } ); --> should'nt a.tabIndex - b.tabIndex work just fine?
[16:50:03] 	(the other code works too and I'll deploy it, this is just aesthetics)
[16:51:40] 	i thought sort made changes to the order of an array based on a negative or positive value being returned
[16:52:12] 	hmm yeah i guess that would work then :)
[16:52:40] 	Yeah that behavior is used precisely because it makes stuff like that work :)
[16:52:47] 	nkomura, adam_miller: Tab fixes deployed, please test
[16:53:08] 	RoanKattouw: en.wp?
[16:53:21] 	All
[16:53:33] 	(Deployed on all wikis that is, just pick any to test on)
[16:53:41] 	k
[16:53:50] 	guillom: great news.  which one?
[16:54:14] 	?? usability matters ??
[16:54:14] *nkomura 	goes to check tabbing
[16:54:41] 	guillom: where is it from?
[16:55:28] 	apparently they're based in Toronto
[16:55:48] 	nkomura: Do we still have the team meeting at 10am?
[16:55:56] 	yes
[16:56:09] 	can you make it RoanKattouw?
[16:56:17] 	Yes
[16:57:40] 	nkomura: So do you approve of my blog post? Can I post it?
[16:57:48] 	Not meaning to pressure you but I'm gonna have dinner soonish
[16:57:57] 	i'm writing comments right now
[16:58:30] 	RoanKattouw: it's not working because the iframe does not have a tabindex
[16:58:49] 	i'm guessing you missed merging the revision where I added that
[16:59:07] 	Looking
[16:59:49] 	You're right, sorry about that
[17:02:10] 	adam_miller: Deployed r62127, try now
[17:02:15] 	k
[17:02:41] 	looks good
[17:03:16] *RoanKattouw 	-> dinner
[17:07:12] 	guillom: that is the UX firm Sue used to work with when she was at CBC
[17:07:25] 	nkomura, oh ok
[17:28:02] 	is IE8 supported on XP?
[17:31:39] *MC8 	launches into an "omg! ie6" tirade
[17:32:15] 	nkomura: yes
[17:32:42] 	do you have that combo handy?
[17:33:45] 	no, on 7. Sorry
[17:33:57] 	np
[17:34:28] 	japanese wikipedians are reporting that pasting doesn't work with IE8 on XP
[17:34:50] 	i could not reproduce the problem in IE8 on Windows 7
[17:42:48] 	nkomura: did you need ie6?
[17:42:59] 	or something tested in ie6 rather?
[17:43:07] 	no ie8 on xp
[17:43:14] 	do you have that combo on your VM?
[17:43:19] 	yep
[17:43:28] 	oh cool
[17:44:00] 	will you check if pasting has issues with all beta features enabled?
[17:44:17] 	japanese wikipedians are reporting that bug and says that it is linked with our toolbar
[17:44:27] 	i can already tell you it does
[17:44:40] 	pasting into the iframe has two issues i identified in IE8
[17:45:22] 	what issue were they reporting?
[17:45:29] 	they can't paste at all
[17:45:50] 	oh haha, no i can paste but it's not perfect
[17:46:01] 	i could paste in IE8 on Win7
[17:46:31] 	brb
[17:47:27] 	back
[17:47:33] 	i know there are issues
[17:47:52] 	but not being able to paste is a drastic one
[17:49:14] 	has anyone from that community reported it recently?
[17:49:31] 	since all of yesterday's code was deployed?
[17:49:55] 	nkomura: what are users unable to paste?
[17:50:02] 	yeah as of last night
[17:50:33] 	nimish_g: it didn't say specifically, but they say they can't copy and paste
[17:50:49] 	so i assume they can't paste anything
[17:51:02] 	is there an bug already about the fact that when you copy a text from the edit window and you paste it somewhere else, e.g. an offline text editor, it adds way too many newlines?
[17:51:22] 	I assume this is a side-effect of something you're already working on
[17:54:07] 	guillom: will you add your comment to https://bugzilla.wikimedia.org/show_bug.cgi?id=22398
[17:54:07] 	i think it is related to this bug
[17:54:23] 	copy+paste is working as far as I can tell
[17:55:15] 	nkomura, done
[17:55:38] 	thanks
[18:03:10] 	TrevorParscal and nimish_g: I'm in R4
[18:03:22] 	walinkg there now
[18:05:58] 	adam_miller: we are ready to meet/talk
[18:06:07] 	RoanKattouw_away: ru back from dinner?
[18:06:24] 	nkomura: conference line?
[18:07:44] 	yup
[18:08:07] 	Sorry guys I only just finished dinner, I'll call in in a minute
[18:10:36] 	nkomura: What's the passcode? Google Calendar is down so I can't look it up
[18:10:42] 	k
[18:10:58] 	25701422
[18:11:08] 	hi RoanKattouw_away
[18:11:24] 	etherpad ftw!
[18:12:03] 	"You have entered an invalid passcode"
[18:12:34] 	sorry
[18:12:46] 	one sec
[18:12:59] 	257017422
[18:13:02] 	and #
[18:14:15] 	did it work
[18:14:19] 	?
[18:16:38] 	http://etherpad.com/BIWzOkAfa0
[18:22:07] 	BTW, etherpad is open source; a local wikimedia version would allow you to do other things, iirc
[18:37:14] 	MC8: we are in the process of setting up an etherpad server already
[18:43:04] 	are we?
[18:46:46] 	http://etherpad.com/BIWzOkAfa0
[19:08:39] 	TrevorParscal, nkomura: So I'd like to deploy the specialchars fix and post the techblog thing, I can haz approval?
[19:11:31] 	asking....
[19:16:23] 	Yes
[19:16:34] 	let's test the software first
[19:16:37] 	on test
[19:16:50] 	and then write something up
[19:17:11] 	and check the top 10 wikis to see if they need the EditTools.js patch as well
[19:17:17] 	patching where appropriate
[19:17:21] 	dewiki doesn't at least
[19:17:23] 	then we will push to the cluster
[19:17:28] 	OK
[19:17:30] 	and publish the blog post
[19:17:41] 	you write it up, I will review it (the blog post)
[19:17:43] 	OK so I have to push the edit.js fix to test now?
[19:17:51] 	yes
[19:18:01] 	Yeah I'm talking about the *other* blog post I've already drafted
[19:18:03] 	Will forward it to you
[19:18:22] 	awesome
[19:18:32] 	Forwarded
[19:28:13] 	TrevorParscal: Sadly I can't push this out on test alone, because it changes JS
[19:28:23] 	And that's hosted on bits.wikimedia.org
[19:28:27] 	Oh wait I *can*
[19:28:36] 	By only changing the style version on test
[19:28:39] *RoanKattouw 	does evil laugh
[19:28:54] 	I don't know about this bits thing
[19:29:19] 	it makes me feel sort of... annoyed
[19:30:12] 	Trust me, it's a good development
[19:30:28] 	It just changes our infrastructure and we have to find our way around that, that's the nature of change
[19:32:07] 	TrevorParscal: OK try test now
[19:32:13] 	I don't mean to be resistant to change, but I keep getting this annoyed feeling every time we interact with bits, similar to that of someone poking me repeatedly in the shoulder to get my attention despite my best efforts to ignore them
[19:32:35] 	I understand
[19:32:36] *TrevorParscal 	is testing
[19:33:09] 	RoanKattouw: working great!
[19:33:20] 	Nice1
[19:33:28] 	Will push cluster-wide on your mark
[19:33:46] 	RoanKattouw: I don't quite like how you say "replacement of the textarea on the edit page with a rich text editor"
[19:33:53] 	it's a bit of a misstatement
[19:34:03] 	it makes it sound like we are allowing rich text editing
[19:34:08] 	which we are, of course, not
[19:34:20] 	Fair point
[19:34:25] 	I was just copying the Wikipedians' term
[19:34:35] 	I'll just call it a content-editable iframe then
[19:34:46] 	yes, well that misunderstanding is causing allot of the problems imo
[19:34:52] 	yes, that sounds much better
[19:35:01] 	it's a fairly technical post
[19:35:12] 	"with a content-editable iframe (called a "rich text editor" by some users)"
[19:35:20] 	Also changed title to "Iframe bugs"
[19:35:44] 	maybe, rather than the parenthesis
[19:36:39] 	You mean a comma?
[19:36:44] 	yes
[19:36:46] 	sorry
[19:37:05] 	spacing or quotes might have helped there
[19:37:57] 	Hm well I was using parentheses to kind of diminish the importance of the statement
[19:38:07] 	Because we kinda wanna be communicating that rich text editor is a misleading term
[19:38:32] 	TrevorParscal: Also, can I push the edit.js fix out to the rest of the cluster?
[19:38:50] 	yes
[19:39:01] 	Cool
[19:40:08] 	RoanKattouw: yes, maybe even say "which some users have been incorrectly referring to as a rich-text editor"
[19:40:36] 	or "mistakenly" rather than "incorrectly"
[19:40:53] 	'with a content-editable iframe (incorrectly referred to as a "rich text editor" by some users).'
[19:40:57] 	s/incorrectly/mistakenly
[19:41:04] 	edit.js fix live cluster-wide
[19:41:06] 	sure
[19:41:08] 	awesome
[19:41:32] 	you just fixed bug 22413
[19:42:12] 	How the hell is that?
[19:42:16] 	bug 22394 is just about done
[19:42:22] 	nimish_g: have you commited it yet?
[19:42:23] 	Oh so we didn't kill the toolbar, it killed *itself*?
[19:42:31] 	RoanKattouw: yes
[19:42:43] 	because it bugged out when getting a selection object on a hidden textarea
[19:43:13] 	TrevorParscal: yes, it's in SVN, but I'm testing on IE and it's not working
[19:43:13] 	22393 as well
[19:43:46] 	Yeah that was the one I intended to fxi
[19:43:48] 	RoanKattouw: can we check the top 10 wikis to see if any copied the en.wikipedia EditTools stuff?
[19:45:01] 	On it
[19:45:10] *RoanKattouw 	figures out what the top 10 wikis are
[19:45:20] 	TrevorParscal: Also e-mailed a new draft of the tech blog post
[19:45:40] 	lookin
[19:45:58] 	bug 22398 doesn't have a fix ready
[19:46:01] 	we still have no clue
[19:46:15] 	OK
[19:46:27] 	the fix we have for 22394 doesn't fix it
[19:47:04] 	adam_miller: what are you currently tackling?
[19:47:54] 	listing out a few weird copy/paste things in IE8 i noticed this morning
[19:48:00] 	TrevorParscal: That's why it says 'needs more testing', want me to rephrase?
[19:48:12] 	anything code wise I can help tackle?
[19:48:21] 	it says fix ready - that's not true
[19:48:24] 	we have no fix
[19:49:22] 	RoanKattouw: have we been seeing any issues with bind and delayedbind being bound to the same events?
[19:49:37] 	that comment was due to my confusion of mixing up bugs
[19:49:54] 	adam_miller: do you know wnaything about bug 22428?
[19:50:04] 	looks like it needs confirmation that it is indeed feixed
[19:50:11] 	TrevorParscal: From what Adam told me the paste issues are at least partially fixed
[19:50:26] 	there's a revision (RoanKattouw do you know which one?) that should be fixing it
[19:50:47] 	Was he the one working on that?
[19:50:50] 	TrevorParscal: that seems fixed in chrome, and it occasionally works in safari, but it's the rarity
[19:51:00] 	which version of safari?
[19:51:02] 	It's Nimish's rev
[19:51:08] 	4.something something
[19:51:09] 	we are only supporting 4 atm
[19:51:18] 	so supporting 3 right now is not a priority
[19:54:13] 	Oh bug 22428, no idea which rev that is
[19:54:24] 	Maybe it is Nimish's, no idea
[19:54:56] 	hmm
[19:55:54] 	line breaks picked up right when I tested in Safari 4, before any fixes and after
[19:56:10] 	WTF pt is spelled Portuguese with a u after the g?
[19:56:10] 	RoanKattouw: so, you mentioned you know what needs doing to fix cursor position after encapsulateSelection
[19:56:20] 	Well sort of
[19:56:29] 	I'm gonna do it right after I finish this
[19:56:47] 	awesome
[19:58:04] 	ooh, it's linebreaks in formatted text
[19:58:43] 	?
[19:59:13] 	adam_miller: are you looking for something to poke at, or are you plenty busy?
[19:59:43] 	looking for something to poke at
[20:00:01] 	the loading stuff is low priority
[20:00:11] 	but probably a good thing for you, it needs design TLC
[20:00:18] 	basically, when the wikieditor loads
[20:00:33] 	TLC?
[20:00:34] 	we cover the textarea with an opaque div
[20:00:40] 	Tender Loving Care
[20:02:07] 	and the div has a spinner
[20:02:11] 	maybe "loading" text
[20:02:19] 	ok, that's something that's not currently in place, right?
[20:03:23] 	Nope, it's new
[20:04:33] 	i gotta get some food quickly
[20:04:50] 	brb
[20:07:07] 	TrevorParscal: Can you hold a few minutes?
[20:07:24] 	RoanKattouw, he's gone out for lunch
[20:07:48] 	Dammit
[20:08:22] 	TrevorParscal: I e-mailed you the fourth iteration of my blog post, could you approve it so it can finally go out? We need to get this status update out there before there's nothing to report on anymore
[20:08:47] *RoanKattouw 	complains about the slowness of virtually all review processes
[20:09:52] 	RoanKattouw: i thought the first blog post was good to go
[20:10:25] 	Yeah Trevor had some (valid) comments
[20:10:35] 	k
[20:10:58] 	nkomura: I  forwarded you my latest version so you can look at it too
[20:11:06] 	k
[20:11:16] 	RoanKattouw: you are off in one hour right?
[20:11:21] *RoanKattouw 	glares at whomever changes behavior to behaviour
[20:11:23] 	nkomura: Exactly
[20:13:25] 	Are they getting lunch to eat in the office, or are they eating out?
[20:13:42] 	i think they are bringing back food
[20:13:55] 	RoanKattouw: the blog looks good to me
[20:14:11] 	OK sounds like he'll be back in time then
[20:14:58] 	yeah he should be back shortly
[20:17:22] 	RoanKattouw: is there bug ID associated with the cursor position you talked about?
[20:17:23] 	Cursor position after encapsulateSelection
[20:17:44] 	is the description in EtherPad
[20:17:59] 	Nope
[20:18:05] 	Well, maybe, lemme see
[20:19:26] 	https://bugzilla.wikimedia.org/show_bug.cgi?id=22379
[20:24:53] 	we overlooked it.. it was reported against test
[20:25:47] 	Not overlooked, I knew about it all along
[20:25:52] 	Before Calcey reported it even
[20:25:56] 	I'm looking into it now
[20:26:06] 	i didn't
[20:26:34] 	adjusting the importance too high
[20:26:57] 	is the fix in the queue?
[20:27:44] 	I'm working on it
[20:28:31] 	the subject of the bug and the description do not match
[20:28:42] 	the subject line ready non-IE, while the report is against IE
[20:29:47] 	Read more carefully: it says the cursor moves to *after* the 
 (correct behavior) on IE and *before* it (wrong behavior) on all other browsers
[20:29:57] 	Calcey *thought* this was an IE bug but it's really all the others that are wrong
[20:32:53] 	hm, i'm not sure if this is the bug case to work from
[20:32:58] 	it is related
[20:33:12] 	It's the same issue essentially
[20:33:19] 	The real bug is broader than Calcey's report of course
[20:34:12] 	right
[20:34:46] 	the line break function has a different cursor position
[20:35:44] 	And so does everything else BTW
[20:36:00] 	The difference described is not specific to the line break button but applies to all insertions
[20:36:21] 	some formatting tasks such and bold and italic
[20:36:39] 	need the cursor to be at the end of the formatting
[20:36:49] 	not at the beginning of the new line
[20:37:37] 	Exactly, at the end of what was inserted
[20:37:52] 	And it so happens the line break button inserts a new line :)
[20:41:26] 	i still see it as a different cursor position,
[20:41:35] 	but i know we are talking about the same
[20:42:04] 	parutron is going through the use case to make sure that we do not follow what calcey is suggesting as expected behavior
[20:42:35] 	Yeah so what I'm shooting for is putting the cursor after whatever was inserted, or selecting part of the inserted text in certain scenarios
[20:43:22] 	Certain scenarios means: if you have no selection and click something that has an example text (e.g. bold button), it'll select the example text after insertion
[20:43:30] 	So it'll insert '''Bold text''' and select Bold text
[20:43:51] 	hi
[20:45:32] 	hi werdna
[20:45:46] 	what time is it in Australia?
[20:46:25] 	RoanKattouw: that sounds right
[20:46:44] 	7:46 AM
[20:46:47] 	7:46 am right?
[20:47:05] 	I like my position, I'm right between PT (-9) and AEDT (+10)
[20:47:49] 	werdna: you are up early
[20:48:27] 	Whoa that's a short night dude
[20:48:32] 	You were up till after midnight
[20:48:36] 	omg
[20:48:41] 	that was the longest wait for food ever
[20:48:44] 	i'm back
[20:48:44] *RoanKattouw 	lassos TrevorParscal
[20:48:46] 	how's it going?
[20:48:57] 	i can haz status update?
[20:49:11] 	TrevorParscal: I'm planning to go to sleep in 15 mins, can you review my latest blog post draft?
[20:49:12] 	nkomura, / parutron Bolt/Peters have decided not to submit a proposal for our study
[20:49:22] 	(fyi)
[20:49:28] 	RoanKattouw: just did, it looks great
[20:49:29] 	Too bad, I liked those folks
[20:49:30] 	they say they're too busy
[20:49:35] 	I give it the "green light"
[20:49:35] 	TrevorParscal: OK publishing it then
[20:49:54] 	nkomura: I've started waking at 7, so that I can actually get work done before lunchtime
[20:50:14] 	guillom: for the multimedia study, or the march usability study?
[20:50:16] 	starting work at 10:30 sucked.
[20:50:25] 	TrevorParscal, multimedia study
[20:50:27] 	guillom: did they say why?
[20:50:41] 	werdna: Try waking up at noon and starting work at 2:30 :)
[20:50:47] 	*that* sucks
[20:50:47] 	werdna: good strategy
[20:50:59] 	Not much of an alternative though if you went to sleep at 2:30 the previous night
[20:51:21] 	Tech blog post published: http://techblog.wikimedia.org/2010/02/iframe-bugs/
[20:51:24] 	RoanKattouw: i want to make sure your sleep schedule can get back to something sane
[20:51:24] 	nod :)
[20:51:34] 	i went to bed 1am but manged to get up at 5:30
[20:51:35] 	nkomura: Could you link that blog post on [[en:WP:VP/T]]?
[20:52:24] 	When I lived in SF I was sleeping from 2 till 8 5 days a week
[20:52:25] *guillom 	needs his 9 to 10 hours of sleep
[20:52:28] 	I was perpetually tired.
[20:52:41] 	RoanKattouw: thanks for chekcing the other wikis
[20:52:51] 	styed up so late to try to catch the IM peak times in Sydney
[20:53:11] 	Ugh
[20:53:23] 	RoanKattouw: it'd be better you do it
[20:53:35] 	I would be tired all the time with 2-8 as well, and I wouldn't survive 1-5:30 multiple times in a row
[20:53:43] 	OK
[20:53:43] 	you have power to do that ;)
[20:54:20] 	It's nice to have PT business hours in the morning now
[20:54:39] 	though, London wasn't too bad last year, London business hours finished at midnight
[20:55:56] 	adam_miller: how's it going over there?
[20:56:45] 	werdna: 1am presumably? Business hours finish at 2am for me, and I'm 1 hr ahead of London
[20:57:01] 	good, inserting a div after the textarea on line 1175 of wikieditor.js
[20:57:04] 	seemed like the right place
[20:57:45] 	RoanKattouw: during daylight savings time it was midnight, I think
[20:58:08] 	That's only possible if SF has DST while London doesn't
[20:58:12] 	As they're on the same hemisphere
[20:58:37] 	RoanKattouw: time check, two minutes before you go to bed
[20:58:40] 	adam_miller: good choice
[20:58:43] 	Yeah
[20:58:51] 	mmm, I don't remember the details, but I seem to remember that being the case at some point. Maybe I'm wrong
[20:59:19] 	It was probably for a month or a week or so
[20:59:24] 	The DST switch dates aren't equal
[20:59:30] 	yes.
[21:00:29] 	http://en.wikipedia.org/wiki/Wikipedia:Village_pump_%28technical%29#Iframe_issues_status_update
[21:01:23] 	RoanKattouw: awesome
[21:01:25] 	I have a meeting
[21:01:33] 	you need to go to bed
[21:01:34] 	:)
[21:01:42] 	i will have my laptop open
[21:01:46] 	if there's any urgent stuff
[21:01:51] 	Yeah I'm off
[21:02:01] 	I have a half-written fix for the encapsulateSelection thing that I'll finish in the morning
[21:06:02] 	good night RoanKattouw_away, good luck with the test tomorrow
[21:06:09] 	Thanks
[21:06:14] 	I'll check back in after I shower
[21:07:24] 	We should arrange to have RoanKattouw_away banned from IRC for certain hours :P
[21:08:03] 	hehe
[21:08:10] 	he would still commit :)
[21:09:16] 	is the fix for preserved formatting staged?
[21:13:47] 	RoanKattouw_away: probably want to update the blogpost diff link for charinsert-fix to http://en.wikipedia.org/w/index.php?title=MediaWiki%3AEdittools.js&action=historysubmit&diff=342986278&oldid=340525133
[21:29:31] 	guillom, flipzagging, and mdale, shall we meet?
[21:29:45] 	yep
[21:30:39] 	yea
[21:32:58] 	is there a quick way to disable the new editor ?
[21:33:10] 	Prefs? It is on enwn
[21:33:19] 	actually, i mean in code.
[21:33:34] 	there are some autoedit scripts that seems to have broken.
[21:34:07] 	was wondering if there is a way to tell disable the editor on a page trough javascript or something.
[21:34:56] 	and i can't rewrite all of them to use the editapi. I have time, but not THAT much :D
[21:45:46] 	thedj: You can probably hack it up with $j.fn.wikiEditor = function() { return $j(this); } no guarantees
[21:46:00] 	Another
[21:46:05] *werdna 	eyes RoanKattouw 
[21:46:05] 	urk.
[21:46:39] *RoanKattouw 	is just doing a quick check-in between showering and going to sleep
[21:50:21] 	RoanKattouw: hey. so the code for pasting seems to work in IE, FF and Safari now
[21:50:51] 	should I update prototype with it?
[21:56:09] 	TrevorParscal: just committed my div cover shiz
[21:56:33] 	sweetness
[21:56:34] 	checkin
[21:56:34] 	take a look and let me know if that's inline with your ideas of how it should work?
[21:59:56] 	adam_miller: Can we maybe insert the covering div earlier, like, as early as possible?
[22:00:47] 	RoanKattouw: like line 186?
[22:01:13] 	oh or maybe 216...is that the earliest it could be?
[22:01:25] 	Not necessarily there, just before we do half a dozen .wrap()s on it
[22:03:15] 	ha, so just move it before the wraps?
[22:03:43] 	Yeah
[22:03:53] 	You know skipping over a few JS assignments is no biggie
[22:04:02] 	The real problem is doing all this DOM manip, that takes time
[22:04:13] 	Causes the textarea to flicker in and out of existence, that kinda stuff
[22:06:15] 	parutron: If I've interpreted your bug comment correctly, you just made my life a whole lot easier :)
[22:08:32] 	TrevorParscal: any thoughts on the loading cover?
[22:09:26] 	the great northwest snow machine has started up again, so i want to head home soon to finish my workday from there
[22:10:43] 	looks cool
[22:10:48] 	go home
[22:10:56] 	one more thing before i go
[22:11:00] 	yes?
[22:11:01] 	what can i work on when i get home?
[22:11:13] 	this looks great!
[22:13:50] 	cool, well maybe i'll just check in when i get back home
[22:14:01] 	if you think of another task in the meantime shoot me an email
[22:14:43] 	until then
[22:25:43] 	philippe from strategy just showed me how the line break is stripped from pasted text
[22:27:25] 	nimish_g: if you have the fix for preserved formatting to stage
[22:27:29] 	please go ahead and do that
[22:27:47] 	ok, which sandbox?
[22:28:19] 	to deployment
[22:29:38] 	does anyone know what ?? Terms are Net/30. ?? means?
[22:30:02] 	yes-ish
[22:30:37] 	"Terms" is on an invoice, how you have to pay them
[22:30:45] 	30 means 30 days -- the invoice is due in one month
[22:30:50] 	I don't remember what net is.
[22:31:10] 	nkomura: I think it still needs to be reviewed and tested before being deployed
[22:31:11] 	net = tax included?
[22:31:19] 	It just means the total outstanding amount
[22:31:22] 	http://en.wikipedia.org/wiki/Net_30
[22:31:30] 	ah, thanks
[22:31:32] 	nimish_g: deployment is the staging area
[22:31:57] 	which runs the deployment code
[22:47:30] 	nkomura: unfortunately Roan ran the svn stuff as himself, so it won't let me stage on it
[22:47:34] 	I'll send him an email
[22:52:12] 	can someone override that as root?
[22:52:28] 	no, it's a security thing
[22:52:39] 	supposed to be a feature
[22:52:44] 	on the prototype, or the actual site?
[22:52:55] 	on prototype
[22:53:03] 	Roan set it up that way for some good reasons, but we need to find a way for anyone to update it
[22:53:14] 	you can still update the prototype sites though nimish_g
[22:53:21] 	right
[23:02:27] 	TrevorParscal: what's your take on the following bug?
[23:02:28] 	https://bugzilla.wikimedia.org/show_bug.cgi?id=22428
[23:02:41] 	philippe from strategy reported something similar
[23:03:08] 	we will not be supporting the syntax highlighting portion of this case
[23:03:12] 	nkomura: it's almost definately related to the arbitrary line breaks in pasted text we are seeing in other browsers
[23:03:17] 	seems to be the oposite issue
[23:03:40] 	right
[23:03:58] 	so this is different from the problem we fixed yesterday
[23:04:07] 	yes
[23:04:14] 	it's a paste thing
[23:04:23] 	k
[23:04:30] 	so that's the tough one
[23:05:20] 	yes
[23:11:01] 	adam_miller: you home now?
[23:14:16] 	yeah! with a now well-stocked fridge.
[23:16:24] 	TrevorParscal: got your email, i'll start looking at the undo stack
[23:20:59] 	hey adam_miller, is your work with the iframe loading "done"?
[23:21:11] 	or is it a wip?
[23:21:32] 	parutron: "done"
[23:21:41] 	i've got a couple issues still
[23:22:19] 	i still see distinct steps:  text frame appears, text frame with cover + spinner appears, iframe appears
[23:22:27] 	i'm hoping the first two can be merged
[23:22:46] 	and the frame/div covering the text frame not be transparent
[23:22:55] 	ooo i can handle the last one :)
[23:23:25] 	hmm, but the sequential load?
[23:23:38] 	making the cover appear instantly would require changing the actual markup that php is outputting
[23:23:54] 	which I'd have to ask TrevorParscal or Roan if we can actually do
[23:24:18] 	TrevorParscal: you around?
[23:24:58] 	you are not at your desk.
[23:25:01] 	where are you hiding?
[23:25:06] 	i kinda doubt that we can do that though
[23:25:11] 	or that it'd be smart to do
[23:25:24] 	if we can't, then i think we'll be making a bad problem worse
[23:25:38] 	but let me see what it looks like with the 100% opacity first
[23:27:11] 	one more thing - can we keep that div/cover to only the text area.  i.e. NOT the toolbar
[23:36:21] 	parutron: http://i48.tinypic.com/2h2o1zc.png
[23:36:43] 	parutron: possibly....
[23:37:41] 	parutron: adam_miller: the problem is that the javascript is executed after the page is loaded
[23:37:55] 	and we don't know if the editor is on or off until the javascript is run
[23:38:22] *TrevorParscal 	tries to think of scarry hacks
[23:39:47] 	oh no
[23:40:04] 	but hmm, this problem won't be there when we go default, right?  because everyone will have the editor.....
[23:45:56] 	parutron: no, people could still turn it off
[23:46:01] 	it would just be on by default
[23:46:17] 	we may be able to figure out a strange hack though...
[23:46:20] 	I will see
[23:46:23] 	cool.
[23:47:13] 	let me know what happens.  i think the solution adam's working on might be aggravating rather than mitigating the problem....
[23:50:20] 	parutron: i agree, but i'd call it pointless rather than aggravating :)
[23:50:37] 	tee hee
[23:50:38] 	should i remove the opacity for the time being?
[23:50:57] 	let me take a look again quickly
[23:52:12] 	if you can do it in 100% opacity white, just in the edit box (not over the toolbar), lets try it like that in case trevor's hackery doesn't work.
[23:52:15] 	adam_miller: I thought the shade should only be over the textarea
[23:52:47] 	and I thought it should have been opaque as well
[23:56:58] 	TrevorParscal: so just hide it more quickly?
[23:57:42] 	it *is* over the text area, but the css is set to scale height-wise and width-wise so when the other divs are inserted it then covers them too
[23:58:03] 	No, what I'm suggesting is that the shade only cover the textarea
[23:58:27] 	it seems to be really tall
[23:58:40] 	i know what you mean about the TOC, that makes sense
[23:58:43] 	that's a bug i'm about to commit a fix for
[23:58:47] 	k
[23:58:55] 	should it cover the TOC and toolbar until removed?