[13:42:42] is there a plan to/problem with upgrading to jQuery 1.4? [15:44:02] Good morning [15:44:11] Hm I don't think I've ever seen this channel fuller [15:44:32] hi RoanKattouw [15:45:43] RoanKattouw: did you get my MM invite? [15:46:19] The one for this morning's meeting with Mark? [15:46:24] yes [15:47:13] Yeah I did get it but you sent it to rkattouw@wikimedia.org instead of roan.kattouw@gmail.com , only the latter is linked to my Google Calendar [15:47:28] So I couldn't confirm my attendance using GCal [15:48:02] k [15:48:21] i'll try to remember to use your gmail account [15:48:56] It's not a big deal if you don't, you just won't get the confirmations from me [15:51:25] k [16:54:43] Are we ready to enable dialogs to the beta users today? [17:22:09] RoanKattouw_away adam_miller - good morning [17:22:18] Morning TrevorParscal [17:22:28] how's it going? [17:22:49] good, how are you? [17:23:14] great - I'll be moving to SF this week, so things will be progressively less sane as the week goes on [17:23:44] adam_miller: we will soon be switching our focus to the features we plan to go default with, so that will be a nice change of pace [17:23:45] whoa, where at? [17:23:58] 6th and Brannan [17:24:08] any progress on those dialog bugs? [17:24:44] are you moving into an apartment in a tall building? [17:24:54] I'm looking at the IE insert stuff now [17:25:25] bug 22877 [17:25:45] TrevorParscal: I'll see you at R5 a few minutes before 11am after I am out of a meeting [17:26:14] adam_miller: I'm moving into a loft - 5 story building, and I will be on the 5th floor [17:26:40] adam_miller: I know a little about that, I was poking at it a few times in the last several days [17:26:42] TrevorParscal: I'm working from my grandparents house m-w this week, so I'll be spontaneously available those days. probably a lot of working at night after they're asleep :) [17:27:08] TrevorParscal: does your loft have full walls or partitions? [17:27:09] ha ha - which starts at like 7pm? [17:27:30] somewhere around 9-10pm usually [17:27:50] the upstairs is totally open, and there's just a railing where it overlooks the downstairs [17:27:51] they're not THAT old yet [17:28:06] and the downstairs has a room with a sliding wall [17:28:17] sounds awesome [17:28:28] I'm pretty excited [17:28:44] i bet! [17:28:56] so what do i need to know about IE insertion problems? [17:29:10] adam_miller: the bug you are working on has to do with IE forgetting the cursor position when you click outside of the textarea - which has to do with selection happening on the document not the element [17:29:37] I added some e.preventDefault() calls to go along with the return false calls we were using in event handlers [17:30:10] (in IE you have to do both - go figure) which helps with things like the heading drop-down since it no longer steals focus and thus selection is not lost [17:30:56] but if you use a dialog, we can't return false on everything [17:31:17] so we have to figure out a way to save/restore the cursor position when a dialog is opened/closed [17:31:38] the problem here has been that IE doesn't properly count newline characters [17:32:13] even if you account for the \r\n IE will ignore selecting newlines, and not count them in it's offsets that it returns [17:32:17] there are some dirty hacks [17:33:22] but the cleanest thing I've seen so far is to cache a copy of the range object [17:33:26] Sounds like this relates to all of my severity: Major bugs [17:34:21] drupal has some JS code that caches a duplicate of the range object -> http://msdn.microsoft.com/en-us/library/ms535872%28VS.85%29.aspx [17:34:54] http://groups.drupal.org/node/1210 [17:35:40] at first I wanted to use the textSelection plugin, so I started thinking of how to add a save/restore selection functionality to it [17:35:59] save would return an object and restore would take that same object [17:36:27] but then I got distracted with other things - so.. that's where we're at [17:36:29] I thought getCaretPosition included these \r\n weirdness hacks? [17:36:30] was that helpful? [17:36:43] yes sir [17:36:56] AFAIK getCaretPosition(true) or whatever that parameter is should return a tuple of the real positions of the start and end of the selection [17:37:03] Unaffected by IE's skipping of \r\n [17:37:59] context.$textarea.textSelection( 'getCaretPosition', { 'startAndEnd': true } ); [17:38:37] RoanKattouw: however, setting the selection in IE was still problematic [17:38:47] How's that? [17:39:15] it would identify the position correctly, but then set it to a position less than what it should be [17:39:30] Interesting [17:39:37] aka, getCaretPosition works well, setSelection seems broked [17:39:38] So setSelection() is affected by the same \r\n bug? [17:39:50] seems to be the case [17:39:54] Then presumably we can just port the \r\n hack used by getCaretPosition to setSelection [17:40:31] I'd do it but there's other things on my list right now [17:40:46] what sort of things? [17:40:55] *TrevorParscal expects long list [17:41:03] Eat dessert - attend meeting - attend other meeting - figure out iframe borkage on live site reported by Nimish over the weekend [17:41:15] all good things [17:41:18] I just worked through a queue of some 60 e-mails piled up over the weekend [17:41:26] adam_miller: do you feel like you have enough to go on now? [17:41:27] Another 150ish to go in mediawiki-cvs and wikibugs-l [17:41:28] guillom: are you done with your recruiting? [17:41:41] parutron, yes, I turned it off Saturday morning [17:41:41] I will be in a meeting from 11 - 12, then grabbing lunch [17:41:42] i'm gonna get the remote recruiter going for our remote study [17:41:44] woo hoo [17:41:44] thx [17:41:59] guillom: how'd it go? [17:42:42] parutron, very well on the English Wikipedia, not that well on Commons, but we don't need that many people from Commons [17:42:45] TrevorParscal: yeah I believe I do. I'll hit you up if i've got more questions [17:44:03] cool [17:54:13] Hey TrevorParscal did you notice http://www.mediawiki.org/wiki/Special:Code/MediaWiki/63925#c6275 has an interesting author? [17:54:21] I didn't know non-coders could comment on revisions ^^ [17:55:11] *RoanKattouw yells at office wireless [17:55:19] Hey TrevorParscal did you notice http://www.mediawiki.org/wiki/Special:Code/MediaWiki/63925#c6275 has an interesting author? [17:55:21] I didn't know non-coders could comment on revisions ^^ [17:56:01] yes - sorry about that [18:06:50] RoanKattouw: I think you mean an "interested author" and not an "interesting author" [18:07:18] GerardM-: No I did not [18:07:22] Usability 2 is a test account [18:07:31] Trevor forgot to log out of it before leaving that comment [18:07:49] Ah ... :) [18:08:06] any news on the Bengali characters ? [18:10:07] News is I'm crazy busy [18:10:41] I'm in meetings all the time today, I'll probably have time to poke at the Bengali chars later this week [18:17:12] cool.. thanks [19:30:37] nimish_g: where'd you gooooooooooo??? [20:19:56] nimish_g: yaaaaaay. updates are on test [20:19:59] and we look good to go! [20:20:14] am gonna head to get some lunch, but is ready to go live whenever you're ready. [20:20:21] the rate from last time worked wonderfully [20:21:01] ok, cool [20:21:13] parutron: how long do we want it running? [20:21:19] like until what day? [20:21:21] 3 days. [20:21:29] wednesday march 24th [20:21:39] i will likely give you a direct kill -9 signal. [20:21:40] tee hee [20:21:55] but generally, i anticipate we'll be done end of the day wednesday [20:22:07] if we crank it up a bit wed. [21:12:11] parutron: ok it should be live on enwiki now [21:12:20] woo hoo [21:12:29] if you wanna check that we're getting recruits =) [21:12:38] clearing my cookies and hitting "random page" [21:12:52] RoanKattouw: did you see anything re: no iframe? [21:13:19] nimish_g: I haven't had time to check, sorry [21:13:30] Fell through the cracks between uni, meetings and e-mail backlog [21:14:20] email? is that that newfangled fancy thing where you get letters inside your glowing computer-box? [21:14:33] I'm gonna go to bed soonish but tell Tom that if he updated to a sufficiently recent version of plugins.combined.min.js (I think he did, looking at the logs) he probably just needs to bump the $wgStyleVersion appendix (i.e. the $wgStyleVersion .= "foo" line) in CommonSettings.php [21:15:03] hm ok [21:15:53] Varnish needs that, otherwise the fix will just work in Europe [21:17:07] 4.5 recruits per hour and counting! [21:17:10] thanks nimish_g [21:17:16] you are radical [21:18:13] question: from localization point-of-view, vecor and the usability initiative are different modules right? [21:18:27] Yes [21:18:32] Vector being in core, UI being an extension [21:18:41] vector's a skin, Usaibility init is an extension [21:18:48] In fact, UI is divided into many submodules such as Toolbar, Toc, Preview, etc. [21:19:59] thanks [21:20:22] GerardM- and siebrand are making the localization completeness summary for us [21:20:29] RoanKattouw: the style version is a string like "z51"...does adding 1 to that make sense? [21:20:46] Just increment the number [21:20:50] So 51 -> 52 [21:22:02] right, I'm asking more "what's that the style number for" ...it can't be for the combined.min file b/c we update that *alot* [21:22:17] there is one overarching Usability Initiative thing that includes all the submobiles [21:22:36] This is what SIebrand created in the current status report\ [21:22:48] nimish_g: No no Varnish needs that extra number because it's aggressive [21:23:18] http://translatewiki.net/wiki/User:GerardM/usability-stats and it does include the vector skin [21:23:55] so we're forcing it to invalidate the cache? (but I'm pretty sure all 'edit' operations bypass our cache and shift-refresh invalidates the user's cache) [21:26:45] This is about Varnish cache for static JS/CSS files [21:27:04] It's similar to Squid but for JS/CSS [21:27:12] aah ok [21:27:15] So it's not purged on edit and Shift+Refresh doesn't work [21:27:20] Only thing that does work is that number [21:28:19] k gotcha, thanks =) [21:33:22] RoanKattouw: changed it, still no iframe [21:33:41] Hm [21:33:46] I have to get some sleep now [21:33:54] I'll poke at it tomorrow when I have time [21:34:16] ok, we definitely need the iframe functional by Thursday morning for our usability study [21:34:31] goodnight =)