[06:37:12] hi! how do i (let) kill a zombie bot at irc.wikimedia.org? [06:38:31] Kill the associated system process? [06:38:48] Usually you can ghost an account on IRC, but irc.wikimedia.org doesn't allow account registration and has no NickServ, I believe. [06:39:01] If you kill the system process, the zombie bot should disconnect immediately. [06:39:16] If not, it should disconnect after a few minutes, at least (it'll have a ping time-out). [06:39:28] it is an otherwise pretty normal IRC network, so a /kill by the ircop should do it [06:39:53] I doubt it's worth a root's time. [06:42:44] Elsie: i thought, i couldn't find a zombie qcron-service at the toolserver. but i just ran top and saw it, and killed it. thanks and sorry for bothering! :-) [06:43:13] No worries. Glad you solved it. :-) [06:44:17] bye. [06:46:19] Another happy customer.^G [06:46:35] ^X [15:10:11] hey folks. i'm trying to skip a step in Guided Tours if the user is logged in. i'm using this code, but it's not skipping. any thoughts?: [15:10:11] shouldSkip: function () { return mw.config.get( 'wgUserId' ) !== 0; } [15:11:59] Ocaasi: might help to also ask in #mediawiki [15:12:20] closedmouth: thank you, good idea [15:14:21] hmm... i think i figured it out. [16:15:07] apergos, parent5446: hello [16:15:22] hello [16:15:36] Hey [16:16:54] so I saw a commit go by but no merge yet [16:17:04] * apergos looks hopefully at Svick [16:17:05] i'm still working on XML output, it should be done now, i'm just verifying that it works correctly [16:17:06] :-D [16:17:08] ah [16:18:03] that's great news [16:18:09] but from my tests, it looks like the output looks the same as the input [16:18:11] it's about time for a new timetable for the next month too [16:18:22] with two exceptions [16:18:36] 1. long comments with invalid UTF-8 [16:18:55] uh oh :-) [16:19:01] that's a known [16:19:02] 2. (i decided not to include those at all) [16:19:23] ok [16:19:45] so if you grep out the restrictions tag the only lines in a (admittedly huge) diff are those comments? [16:20:12] yeah [16:20:17] nice [16:20:32] this would be a good time to have folks play with it [16:20:54] it may not be efficient but you'll get feedback about the format and any edge cases [16:20:57] though i didn't find a good program for diffing huge files (diff seems to work up to 500 MB on my linux VM) [16:21:05] and second then it can go in the monthly engineering report :-D [16:21:09] (i've been testing on tenwiki and cswikisource) [16:21:19] ah memry, yeah. [16:22:15] well you can toss all restriction and comment tags from both files and cmp them (skipping over siteinfo etc if necessary) [16:22:28] that would at least tell you that everything else was identical [16:23:25] yeah, i think i'll do one more change: options for what kind of dump should be on the output: pages/stub, history/current (input is always the full pages-history) [16:23:44] and then i'll update the format spec and the email [16:23:50] hmm rdiff might work for you [16:24:29] i tried that, its output is not very useful [16:24:41] ok bummer [16:26:04] diffing tenwiki works fine (it's small enough) and on cswikisource i've managed to verify 500 MB/640 MB, so i think that should be fine too [16:26:38] if there are differences I am sure other folks will find them for you :-D [16:26:48] they are annoyingly good at that [16:26:57] :D [16:27:21] apergos: https://meta.wikimedia.org/wiki/Cunningham%27s_Law [16:27:30] i'm okay with that :-) [16:27:34] yep [16:27:43] too which I answer [16:27:52] Hehhhhhh. [16:28:09] http://xkcd.com/386/ [16:28:14] (of course) [16:28:59] so I guess I'll wait for the next month's timeline to show up in order to find out what you'll be working on for tomorrow [16:29:54] * Elsie added the xkcd link to Meta-Wiki. [16:30:02] nice one [16:30:29] i'm not sure how much sense does it make to have a week-by-week timeline, since the one for this month was very inaccurate; but i'll try anyway [16:30:58] well if gives folks an idea of where you are headed [16:31:04] even if there is slippage or you change the order [16:31:09] *it [16:31:18] right [16:31:32] and since most folks probably don't watch our discussions in this channel, that's how they would find out what's going on [16:32:02] Mhm, also it provides a bit of a measure of what was done compared to what was scheduled to be done. [16:32:03] ok, i guess that makes sense [16:33:27] sept 16 is supposed to be end of features and debugging, at that point one is supposed to be doing cleanup, docs, etc [16:33:38] so have that in mind as you get set for the next 4 weeks [16:34:33] ok, i will [16:35:44] if you think there will be too much on your plate to complete, we can chat about it tomorrow (at meeting or earlier), figure out what to prioritize [16:37:37] i think it should be doable, but we can certainly talk about what to do first [16:38:43] anything you want to bring up development-wise, or other? [16:40:38] i'm assuming that i'm okay regarding the official mid-term evaluation? (i think i'll post mine tomorrow) [16:42:54] you are fine [16:43:16] ok, cool, in that case, see you tomorrow [16:43:19] if there was a danger of you being failed you would have heard about it from me long since [16:43:24] See you tomorrow [16:43:30] that's the sort of thing we want to prevent, not let creep up on the student [16:43:38] so... [16:43:39] yeah, i just wanted make sure [16:43:45] understandable :-) [16:43:57] so, happy timetabling and see you tomorrow! [16:44:00] bye [17:14:32] andre__: greg-g: so just to be clear: once Andre goes to GUADEC who is the contact in lieu of andre__? is it Greg? [17:15:05] sumanah, yes, Greg [17:15:15] andre__: great. Thank you both :D [17:15:29] sure :) [17:16:32] Svick: did you forget to submit after +2? [17:17:44] sumanah: "it's me, it's me" /me says in the style of the SNL cheerleader skit [17:19:41] :) [17:27:25] apergos: yeah, i did, thanks [17:27:38] yw [17:33:01] http://en.wikipedia.org/wiki/Wikipedia_talk:Articles_for_creation/Hollow_Bodies_(Blessthefall_Album)?veaction=edit is not loading for me. Is that bugreport-worthy? And if so, how should I report it (that is, should I leave the article in the state that it is, so that is reproducible, or should I work on it risking making it non-reproducable)? [17:33:39] it loads for me, so I dunno [17:33:49] * MartijnH shrugs [17:33:51] I am not logged in [17:33:57] I'll try a few browsers [17:34:09] ok [17:34:10] MartijnH: Check your browser's JavaScript error console. [17:34:24] And you can always provide a permalink. [17:34:33] (Use "permanent link" in the sidebar.) [17:34:36] heh [17:34:51] thanks mz, I feel like such an idiot now [17:35:06] No problem. [17:39:31] yeah, working in firefox. I think an extension balked [17:40:53] some people are going to click hese buttons just to see what they do [17:41:00] like the puzzle piece [17:46:41] that's a good thing, isn't it? [17:52:29] I don't have a proble [17:52:30] m [17:52:38] triagers might, we'll see [17:54:57] meh, it seems to be a userscript that broke it [17:55:26] sorry to keep bothering you, but how do I land in debug mode again? [17:56:07] what debug mode? [17:56:11] you mena adding ?debug=1 to the URL? [17:56:22] mean* [17:56:22] I think so [17:56:47] yes, that's what I meant [17:56:48] thanks [18:01:25] and got it, indeed a userscript [18:25:57] MartijnH: ?veaction=edit on a Wikipedia_talk: page? VE should not be enabled on talk pages [18:26:23] ugh, mislink from me [18:26:28] and funny actually [18:26:31] it's a redirect [18:26:42] but VE picks that up and opens the redirected page [18:27:01] (awesomesauce) [18:27:22] heh, OK [18:27:42] is that by design, or working well by accident? [18:28:28] Working correctly for me [18:28:40] MW follows the redirect and VE goes "oh look, a main namespace page" [18:29:37] yeah, for me as well (which is what I was happy about, just wondered if that was intended, or a happy technological coincidence that wasn't given thought yet) [18:30:25] newest conspiracy theory: someone with access to VE tries to discredit anonymous users by making vandalism with VE as IP on dewiki [18:30:40] eliminate him! he knows our plans [18:30:43] sorry, wrong channel [18:32:18] by the way, now that you're distracted anyway, I'm really bad at this stuff, but the offending js that broke VE for me was $($("div.#content").css('marginRight', $("#mwe-pt-toolbar").outerWidth())); which is probably an abomination against all things good and web [18:32:49] but is was working for me. If someone knows how to fix that from the top of their heads, I'd be really happy (but not that urgent, I don't use it all that often) [18:34:01] greg-g: We need to re-arrange or re-assign next mondays deploy [18:34:23] As I'll be over eastern Europe/Russia in 7 x 24 hours [18:34:27] MartijnH: what is that code supposed to do? [18:34:38] also, what are you trying to fix? (i have no context) [18:34:47] <^demon> Reedy: Need me to do it? [18:35:32] MatmaRex, when the sidebar of the page curation tool is displayed, make sure that it doesn't overlap the content of the page [18:36:18] having that line breaks VE, there is no $("div.#content") in VE [18:36:32] .# ? [18:36:46] "abomination against all things good and web" [18:36:50] Yes [18:36:53] For so many reasons [18:37:13] MartijnH: define "in VE" [18:37:31] also yes, i wonder why it has ever worked [18:37:32] <^demon> RoanKattouw: Oh, I saw this over the last few days. I thought of you. http://visual.ly/if-web-browsers-were-celebrities [18:37:37] I will not deny in any way shape or form that I have no idea what I'm doing when it comes to DOM or CSS selectors [18:38:16] MartijnH: you probably want "div#content", and that should work when VE is enabled as well [18:38:17] MatmaRex, the VE won't load through anymore, it will get to 'greyed out', but JS stops execution because it errors on that line [18:38:38] also, yeah, the"wrapping" $() seems unnecessary [18:39:03] let me try what this does [18:40:12] ^demon: Yeah, I saw that one ) [18:40:12] * :) [18:40:44] MartijnH: how do i get the page curation toolbar to appear? [18:40:53] or is it possible to turn it off, maybe? where? [18:42:06] meh, I'll fiddle with it for a bit, undoubtably violating many if not all gods of sane js [18:42:18] it's not all that urgent, but thanks for the help anyway [18:42:23] much appreciated [18:42:29] MartijnH: $("#content").css('marginRight', $("#mwe-pt-toolbar").outerWidth()) [18:42:32] this should work [18:42:41] i'd test if i could get the toolbra to appear [18:42:46] Reedy: agreed re next monday, in fact, (see email, but) let's not do any deploys next week :) [18:43:30] MatmaRex, you can get it to appear when on en.wp you navigate to a page from Special:NewPagesFeed and mouse over the right side of the screen [18:44:17] like, say, https://en.wikipedia.org/wiki/Jamalpur_Bangar ? [18:44:23] greg-g: Poor twkozlowski only just added the next 12 to the roadmap :p [18:44:40] because there isn't one there [18:44:43] oh well, whatever [18:45:09] thanks for taking the time anyway [18:45:30] Reedy: hehe, yeah, luckily it isn't too bad of a edit to "fix it" (just delete a few lines, renumber, er, 9 of the wmfs, heh [18:45:47] and recalculate dates [18:45:55] with months and numbers and stuff [18:46:35] well, if we just do wmf13 starting on the 15th, then we just renumber the wmf15 down to wmf13, 16 to 14, etc [18:46:52] and delete the lines for wmf13+14 (as numbered now) [18:47:24] * greg-g goes to get lunch, bbian [18:47:26] s/n/b/ [18:49:41] huh? what's happening to the deployments? [18:49:56] another US holiday? :P [18:50:05] ah, or wikimania, i guess? [18:50:09] Wikimania [18:50:15] I'll be in the air in 7 x 24 hours [18:50:34] meh [18:52:18] Is git.wikimedia.org slow for anybody else? [18:52:34] ^demon: ^ is it down? [18:52:51] <^demon> Shouldn't be. [18:53:15] http://www.downforeveryoneorjustme.com/git.wikimedia.org [18:53:37] <^demon> Restarted. [18:53:37] hmm, but for me i'd have to confirm i get a Proxy Error now [18:53:48] <^demon> It was using 500% CPU :) [18:53:54] yepp, 503 [18:53:56] ack, turned into 503 [18:54:17] <^demon> The hell gitblit? [18:55:05] <^demon> Turns out restart only stops. [18:55:08] <^demon> What a silly init script. [18:55:21] restop ?:p [18:55:42] works for me again [18:55:47] <^demon> gitblit's back. Thanks for spotting that StevenW [18:55:56] <^demon> It was running, just too slow to be useful :) [18:56:00] thanks for the quick reply [18:57:38] <^demon> yw [18:58:17] <^demon> mutante: Fyi, this is documented on wikitech and has an init script now. Which is better than being undocumented and having to kick it off manually :) [19:00:03] ^demon: cool, thanks:) [19:31:01] I am getting Connection Refused on HTTPS [19:31:06] HTTP works [19:33:09] https://gerrit.wikimedia.org/r/76556 [19:34:52] aude: That version of DataTypes isn't going to make PHP very happy [19:35:03] DataTypes_VERSION is undefined [19:35:13] oh, ok [19:35:23] * aude looks [19:36:24] ^demon: manybubbles: Your Elastica github git submodule isn't going to play nice on the cluster [19:36:42] <^demon> ¿Por qué? [19:37:12] tin cannae access the internets [19:37:35] Cloning into 'Elastica'... [19:37:35] error: Failed connect to github.com:443; Connection timed out while accessing https://github.com/ruflin/Elastica.git/info/refs [19:37:35] fatal: HTTP request failed [19:37:35] Clone of 'https://github.com/ruflin/Elastica.git' into submodule path 'Elastica' failed [19:37:39] <^demon> Lamesauce. [19:37:40] <^demon> Ok. [19:39:58] Reedy_: https://gerrit.wikimedia.org/r/#/c/76608/ [19:40:56] i can update the core branches again, if that is helpful for me to do [19:41:44] Can if you want :) [19:41:54] ok [19:41:58] I'll then "dark deploy"/sync it [19:42:13] not that that file will actually get loaded, but just in case [19:44:05] <^demon> Reedy_: Redid your change. https://gerrit.wikimedia.org/r/#/c/76029/ [19:45:15] Ooh. When did we start not needing the .git? [19:45:36] gitblit shows them.. gerrit doesn't [19:46:47] huh, no .git? [19:47:58] * Reedy_ tests [19:48:43] https://gerrit.wikimedia.org/r/#/admin/projects/mediawiki/extensions/CirrusSearch/Elastica [19:48:46] git clone ssh://reedy@gerrit.wikimedia.org:29418/mediawiki/extensions/CirrusSearch/Elastica [19:50:24] WFM [20:04:33] aude: Has https://gerrit.wikimedia.org/r/#/c/76483/ been updated too? [20:12:50] looking [20:13:59] Reedy_: yes, it has [20:14:21] I merged and attempted a cherry pick of the other update ;D [20:30:24] I notice that the Farsi Wikipedia uses its own numeral system. Is localizing number systems something that's done routinely, or does it require a huge bunch of programming work for every new system? [20:31:37] YairRand: Looks like around 15 lines of config [20:32:24] https://git.wikimedia.org/blob/mediawiki%2Fcore.git/a06bf8f56aa3820ac90c40c1cefd8c4f0c270ffd/languages%2Fmessages%2FMessagesFa.php#L335 [20:33:52] Reedy_: is the section of that page that I'm supposed to be looking at the "$digitTransformTable" [20:34:00] Yes [20:34:21] Previous repo browsers made it more obvious [20:34:35] ah, thanks. are there ways of getting more complicated number systems in? [20:34:49] other than raw digit-to-digit? [20:36:11] Seems there's around 30 languages with it customised [20:36:16] More complicated how? [20:37:06] things like, separate symbol for "10" [20:37:50] If it was coded [20:37:58] By the look of it, it literally does digit for digit [20:38:18] $s = $this->digitTransformTable(); [20:38:18] if ( $s ) { [20:38:18] $number = strtr( $number, array_flip( $s ) ); [20:38:18] }] [20:38:59] I see. [20:39:30] Thanks for your help. [21:47:31] twkozlowski: https://gerrit.wikimedia.org/r/#/c/72738/ :D [21:50:16] errrm. Why me? :-) [21:52:54] because you're the tech news guy :D [21:58:44] !seen John Du Hart [21:58:45] Did you mean @seen John? [21:58:53] @seen john [21:58:53] mutante: Last time I saw john they were quitting the network with reason: no reason was given N/A at 10/12/2012 3:52:43 AM (290d18h6m10s ago) [21:59:51] heh, silly bot [21:59:54] !seen [21:59:54] Did you mean @seen $1? [21:59:59] !seen del [21:59:59] Successfully removed seen [22:00:12] !seen is Did you mean @seen $*? [22:00:12] Key was added [22:00:16] !seen John Du Hart [22:00:16] Did you mean @seen John Du Hart? [22:00:20] good. [22:00:31] actually, irc nicks can't have spaces. lol [22:00:43] what am i doing. man. [22:00:44] i copied from a Maintainer: line :) [22:01:02] mutante: you probably weren't looking for 'john' anyway, y'know ;) [22:01:13] ping johnduhart.me [22:01:34] !ping is 100% packet loss [22:01:34] This key already exist - remove it, if you want to change it [22:01:57] i added !seen myself and didn't use $*, and now i forgot i had a reason for that [22:02:00] oh well [22:02:37] ;) [22:03:50] MatmaRex: wait wut. [22:03:56] There is an Opera 15?! [22:05:00] wait wut, there is also a Firefox 22?! [22:05:18] What has happened to the good, old X.YZ numbering... :-( [22:06:02] Chrome 27 :p [22:06:05] marketing [22:06:10] marketing happened. [22:06:13] Good Lord. [22:07:06] actually, in opera's case, i dunno [22:07:13] but 15 is basically a whole brand new software [22:07:17] they should call it Opera 2 [22:07:44] why they skipped 13 and 14, i have no idea. i don't think they;re planning to releaseupdated to 12 as that [22:08:09] Though now I think of it, I guess I might have used Opera 9.2+ at some point. [22:09:28] opera 9 was the shit [22:09:33] ah, good times [22:09:44] when box-shadow and rounded corners support was a big thing [22:10:43] (Wee, thunderstorm!)