[17:40:06] *^demon is having a Gyro for lunch, the closest I can come to a Doner in the US :p [17:46:52] jorm, about? [17:46:59] si [17:47:03] Hey [17:47:14] Is there an article assessment meeting thing today [17:47:15] ? [17:47:27] i have heard rumors that there is, but it has not appeared on my calendar. [17:47:38] i think alolita wants to go over the phase 2 designs. [17:47:40] that's what I was thinking [17:47:52] And I know alolita tends to be pretty good about getting invites out [17:48:11] she was ill on wednesday, i believe. [17:50:44] she just came into the office. [17:50:51] Ahh [17:50:58] need to work out if i need to head home fairly quickly :P [17:51:06] Though, seemingly no Trevor or Roan atm [17:51:30] no one is here but me. i doubt any movement on this for a bit. [17:52:02] fair enough [17:52:07] Bloody american slackers ;) [18:40:27] jorm, :D [18:41:01] yeah, i pinged to get a time or something. [18:41:20] Is it worth me texting Roan to find out if he's about? [18:42:16] sure, why not? [18:44:09] Someone should suggest he buys a smart phone ;) [18:45:05] jorm/alolita, roan is having dinner so will be late [18:45:28] reedy: hi! no problem - he can join in when he can [18:45:45] I'm just getting myself sorted (recent pc reinstall) [18:45:55] so got to install a sip client and go and make a drink :D [18:47:33] reedy: no p :-) talk to you shortly [19:07:07] RoanKattouw, you're not even late ;) [19:07:46] Didn't start yet? [19:07:48] 2003 I see in the e-mail? [19:07:51] Mmm [19:08:04] Or alolita hasn't poked me yet :P [19:08:14] we're in 2003. [19:10:13] http://www.mediawiki.org/wiki/Article_feedback/Public_Policy_Pilot/Design_Phase_2 [19:12:14] Is there an Etherpad? Link? [19:12:22] Been posted elsewhere.. [19:12:24] Not sure where [19:12:46] http://eiximenis.wikimedia.org/AFT11152010 [19:12:54] Yay thanks [20:46:46] RoanKattouw, I was right [20:46:52] just changing to array, and the code works [20:47:00] I'll update the documentation [20:47:02] Nice [20:47:11] Just change it to array? [20:47:17] Makes more sense than "string or array" [20:47:39] "X or array of Xes" is a common phrase in RL code [20:48:12] Hey, TrevorParscal or RoanKattouw, did you make any changes to theming? [20:48:32] suddenly getting a lot of RL errors like this: [20:48:34] Warning: filemtime() [function.filemtime]: stat failed for /Users/neilk/Sites/wiki/resources/jquery.ui/themes/vector/images/ui-bg_highlight-hard_100_f2f5f7_1x100.png in /Users/neilk/Sites/trunk/includes/resourceloader/ResourceLoaderFileModule.php on line 346
[20:48:43] I didn't touch anything [20:49:20] flipzagging: Does the containing dir exist? [20:49:21] that file exists and is 0744 for me [20:49:26] ^^ [20:49:31] So it's there but fails to stat? [20:49:33] Clear cache maybe? [20:49:43] which cache? [20:50:25] Are the warnings showing up on pages or in CSS/JS? [20:51:55] hang on, pastebin [20:52:23] anyway, it's JS [20:52:28] pastebin is slow [20:53:34] RoanKattouw: http://pastebin.mozilla.org/853152 [20:54:37] flipzagging: themeing? [20:54:50] as in jquery ui themes? [20:54:57] I don't know, stat failed on a few vector theme images [20:55:05] hmm [20:55:07] or so the error message from load.php claims, because the file IS there [20:55:18] I havn't mess with that in quite a while [20:55:39] Tried Shift+Refreshing that? [20:55:58] yes, several times [20:56:10] anyway, the real problem is that somehow a null object gets passed to jQuery [20:56:39] RoanKattouw: if you can review and deploy reedy's code we can begin category migration. [20:56:53] we should have the new category(ies) as well as the current one for the time being. [20:56:55] jorm, technically, it's not even needed [20:56:58] er, correction [20:57:04] just get him to update the site config, and it would work [20:58:03] Alright, just a second [20:58:07] Or a minute rather [20:58:21] hm, it works fine in Chrome [20:58:25] maybe this is my browser, somehow. [20:58:26] let [20:58:28] err. [20:58:30] flipzagging: $(null)? Where? [20:58:40] Also, those stat failures are really weird [20:58:43] I don't know, Firebug doesn't give me a stack [20:58:50] Maybe perms errors? [20:58:51] Let's use: Article Feedback Pilot, Article Feedback, and Article Feedback Additional Articles [20:58:53] and the error didn't occur on CHrome [20:58:56] (No execute bit on dirs) [20:58:58] Hm [20:59:04] Did the stat error occur on Chrome? [20:59:07] then when everything is moved from Article Feedback Pilot -> Article Feedback, we'll remove the one cateogry. [20:59:35] RoanKattouw: yes, that makes sense, the dirs are non-traversable. [21:00:13] at least in my local copy. Is there an svn diff for perms? [21:00:19] flipzagging: So fix your server setup. Apache needs read+traverse access to the resource files [21:00:22] No, SVN does not control perms [21:00:31] AFAIK [21:00:54] Maybe there's some keyword voodoo, but none that we use [21:01:09] So a checkout would just create the files the normal way, obeying your umask and all [21:01:14] oh no, actually it was traversable before, or I wouldn't have seen anything. [21:02:56] or, some dirs were traversable, maybe not all. chmodding the whole resources hierarchy may have fixed it. [21:09:45] alolita: did you speak with RoanKattouw about the possibility of doing Pending Changes during the UploadWizard dark launch? [21:12:54] robla: hi! i did indeed [21:13:20] robla: RoanKattouw does not think launching simultaneously will work [21:14:04] robla: so we can do PC first (on Tuesday next ) since RoanKattouw would deploy UploadWizard on Wednesday then [21:14:18] sounds reasonable, thanks [21:14:40] robla: :-) [21:16:49] We *cannot* launch things simultaneosly [21:17:00] Things get nasty when two people are in there at the same time trying to get different things done [21:19:47] don't worry about that; hasn't our fundraising income always been proportional to the amount of downtime we have? ;) [21:20:01] Bryan: :-) [21:20:35] jorm: Alright, let me help you now [21:20:53] Deploy Reedy's code, set categories [21:21:28] Reedy: + if ( !count( $wgArticleAssessmentCategory ) [21:21:32] That looks like a syntax error [21:21:48] if ( !count( $wgArticleAssessmentCategory ) [21:21:48] || !self::isInCategory( $title->getArticleId(), $wgArticleAssessmentCategory ) ) { [21:22:05] Right [21:22:08] I'm blind [21:22:10] heh [21:22:25] in words it doesn't make much sense, but didn't see much point renaming the global [21:22:34] Minor nitpick: rename isInCategory() ? [21:22:41] Yeah and rename the global too [21:22:43] Some time [21:25:54] RoanKattouw, I'll just do it all now? [21:26:16] what will count() on a non-array do? [21:26:19] Go bonkers, I'm deploying r76704 solo without that [21:26:24] Return 0 or 1 [21:26:24] lol [21:26:44] and a notice or maybe not [21:26:52] No no [21:26:56] isPageInCategories.. [21:26:56] http://php.net/count [21:27:02] I would use empty() [21:27:05] Explicitly defines count() behavior for all types of vars [21:27:21] Tim actively campaigns against empty() in favor of count() [21:27:42] ok [21:28:00] didn't know that it was defined behaviour [21:28:32] as opposed to the "accidently works" like we're used to with php [21:29:24] Well empty() has well-defined behavior too, but it's defined in a scary way [21:29:38] To quote Tim, it's a cast to bool suppressing all errors [21:29:46] So empty($undefinedVariable) returns true, no notices or anything [21:32:31] <^demon> empty() is almost as evil as @ :p [21:34:44] It includes @, pretty much [21:36:33] <^demon> empty( $unset ) is the same as @( isset( $unset ) && (bool)$unset ) [21:36:38] <^demon> So yeah :p [21:38:55] Minor nitpick: rename isInCategory() ? [21:38:59] To what? [21:39:03] I'm unimaginative [21:39:24] isInCategories() ? xD [21:39:53] Fiine P [21:39:54] :O [21:39:56] *:P [21:40:01] I wasn't quite sure of your objection :P [21:40:25] - 'enwiki' => 'Article_Feedback_Pilot', [21:40:31] + 'enwiki' => array( 'Article_Feedback_Pilot', 'Article_Feedback', 'Article_Feedback_Additional_Articles' ), [21:40:35] jorm: That OK? ----^^ [21:40:48] that's perfect. [21:41:41] !log catrope synchronized php-1.5/wmf-config/InitialiseSettings.php 'Adding categories to $wmgArticleAssessmentCategory' [21:44:50] jorm: Done, please verify [21:46:16] is article feedback addtional articles a hidden category? [21:46:27] Not in my control [21:46:37] *RoanKattouw hands jorm a __HIDDENCAT__ [21:46:43] how do i do that? [21:46:50] it worked, but that's not a hidden cat. [21:47:01] I *think* you put __HIDDENCAT__ on the category desc page [21:47:33] Which I don't think invalidates cache for its members, so you may have to action=purge before the cat is hidden [21:47:44] that's cool. it worked. [21:47:47] and confirm. [22:10:20] hey RoanKattouw [22:10:29] Hey flipzagging [22:10:37] so, is everything a go for deploying core features tomorrow? I'm going to commit one fix, just to error handling [22:15:21] I need to catch up on review for that [22:15:25] Fortunately, I have tomorrow off [22:15:27] (from school) [22:15:32] I'm kinda fried right now [22:15:38] np [22:15:44] Got up 6:15, left 7:15, got back home 7:15pm, now 11:!5 [22:16:02] as long as you think you are unfried tomorrow. if not we can post pone [22:17:13] Yeah I will eb [22:17:20] I'll spend tomorrow catching up [22:17:28] And taking it into my hands to fix what needs fixing [22:17:34] Since you'll be asleep if you know what's good for you [22:19:38] hm [22:20:08] we scheduled it for 10am, didn't we? [22:20:14] Yeah [22:20:19] I sleep in from time to time, but I'm usually up by then [22:20:46] anyway, you should be relaxing / sleeping now, I won't bug you. [22:21:00] Kinda should [22:21:11] Except I'm still reading e-mails 24 hrs old [22:21:53] Yeah you damn right I expect you to be in at 10am :P just saying that CR comments ain't gonna cut it when there's like an hour between you reading them and the scheduled deployment [22:25:40] Ok [22:26:02] I'm committing just one change now, should be uncontroversial, and fixes some things people complained about. [22:27:10] I'm not gonna see it till tomorrow anyway [22:27:23] I won't look at anything before midnight PT