[07:58:13] Hi [07:58:21] I see Score extension got deployed [07:58:53] However for whatever reason I can't make it recognise VALID lillypond notation ( I copied one of the templates from the tutorial pages to test it) [08:00:05] Does score uses a different version of the notation ? [08:00:23] http://en.wikisource.org/wiki/User:ShakespeareFan00/scoretesting [08:02:36] If you are going to enable an extension, the leastyou could do is ensure that it CAN process the notation which the manual for that notation SAYS you are SUPPOSED to use. [08:06:05] Alterantivly can some please explain carefully where the syntax given the tutorial sample is worng [08:06:07] *wrong [08:24:09] hi QCODER00 [08:25:01] Hi Tim [08:25:21] Can you explain why the score extension is having 'issues' in parsing what should be perfectly valid notation [08:25:23] ? [08:25:27] it is WRONG because by default, the extension adds its OWN \score tag [08:25:47] OK how do I fix it? [08:26:21] you can either remove the \score section and the "melody = ", or you can use the "raw" parameter [08:26:46] And how do I do that? [08:26:50] ... http://en.wikisource.org/wiki/Page:The_Army_and_Navy_Hymnal.djvu/32 also has issues [08:27:00] but it doesn't look very nice with that second option [08:27:44] http://upload.wikimedia.org/score/g/a/gaxlbyt3hem4686x7jfy4b3x9nwym1u/gaxlbyt3.png [08:27:49] generated with "raw" mode [08:27:57] probably not what you want [08:28:00] No [08:28:11] sorry I mean probably NOT what you WANT [08:28:30] In the second example, It doesn't like the bar line (which is what the lilypond notation guide SAYS you use. [08:29:20] what bar line? [08:29:37] | [08:29:51] < c g > | < c e > [08:29:54] Won't render [08:30:41] Depsite it seemingly being the exact syntax you are supposed to use [08:31:05] but it works if you wrap it in \relative c {} [08:31:34] OK, but the notes didn't exactly SAY that [08:31:45] what notes? [08:32:39] http://lilypond.org/doc/v2.14/Documentation/learning-big-page#single-staff-notation [08:35:36] "Expressions: every piece of LilyPond input needs to have { curly braces } placed around the input. These braces tell LilyPond that the input is a single music expression, just like parentheses () in mathematics. The braces should be surrounded by a space unless they are at the beginning or end of a line to avoid ambiguities." [08:35:45] In any case I can't get the notes to match up with what the score actually says... [08:36:29] So I'm rverting those edits and waiting until someone develops a way of adding scores that doesn;'t need you to endlessly twaek around trying to understand a pedantic syntax [08:36:46] maybe they figure most people read section 1.2.2 before section 2.1.1 [08:37:21] you know you can use ABC input syntax if you don't like lilypond [08:37:23] i wonder if en.wiki has a quick start help: guide yet... [08:37:46] then it becomes much less pedantic [08:38:03] Nah [08:38:24] It's easier to let someone that can work with the syntax do it [08:38:28] it seems like what you're asking for is already done [08:38:59] Well the score extension is delpoyed... [08:39:06] with ABC support [08:39:09] but it's not exactly friendly to use [08:40:01] gtg [08:40:08] I was using lillypond syntax (or attempting to ) because eventually the scoring needs to support the lyrics addition [08:40:19] I'm not sure ABC does that [08:42:33] ON a technical note , sometimes when previewing I get the raw text, and not a rendering [08:42:53] I assume this is by design , although it makes checking the scoring somewhat harder [08:45:05] http://en.wikisource.org/wiki/Page:The_Army_and_Navy_Hymnal.djvu/32 - I put the bar line in - but lillypond seemingly ignored it [08:46:49] Is there a method for forcing a bar rendering at the end of a pick-up measure? [08:47:15] (the pickup being 1/4 long as opposed to 4/4)? [09:37:01] Lydia_WMDE: more Wikidata i18n problems on https://translatewiki.net/wiki/User:Lydia_Pintscher_%28WMDE%29 [09:38:12] Nemo_bis: thanks - looking [09:40:14] Nemo_bis: i'll poke people about them again [09:41:56] Lydia_WMDE: Danwe was nice enough to reply on some of them a few days ago, but not on most [09:42:04] I'm not sure if there's some missing information on either side [09:42:07] Nemo_bis: yeah he replied to the ones he could [11:09:54] Lydia_WMDE: could you add the new list to https://meta.wikimedia.org/wiki/Mailing_lists/Overview#MediaWiki_and_technical please? [11:10:15] And also Gmane if it's not mirrored yet, it helps search etc. http://gmane.org/subscribe.php [11:10:41] Nemo_bis: ok - might take a bit until i get to it though [11:11:10] oki [13:50:05] Nemo_bis: is there a reason i got the rename message like a million times? :) [14:33:13] matanya: yes, sorry, it was an unlucky day with esams quite disrupted [14:33:40] oh, i see, no prob, just wondering [14:34:11] matanya: https://meta.wikimedia.org/w/index.php?title=User_talk%3ANemo_bis&diff=5459431&oldid=5459258 [14:34:19] you're completely right, it was a mess :( [14:58:32] Hi all, any idea why the # of categories under "https://en.wikipedia.org/wiki/Basketball" is 5 but the latest database dump has 16 in categorylinks table ? [15:01:17] pmenon: Hidden categories: https://en.wikipedia.org/wiki/Basketball?action=info [15:02:32] marktraceur: how does one determine if a given category is hidden on a given page using the categorylinks table? [15:03:40] You can't [15:04:12] You'll have to check if the category contains the hidden magic word I think [15:04:53] well ... that isn't good [15:05:54] pmenon: You can with the api.. [15:06:03] clprop - Which additional properties to get for each category [15:06:03] sortkey - Adds the sortkey (hexadecimal string) and sortkey prefix (human-readable part) for the category [15:06:03] timestamp - Adds timestamp of when the category was added [15:06:03] hidden - Tags categories that are hidden with __HIDDENCAT__ [15:06:18] prop=categories [15:06:34] If you have a copy of the page_props table you could as well [15:10:53] SELECT pp_value FROM page_props JOIN page ON page_id=pp_page WHERE pp_propname = 'hiddencat' AND page_namespace = 14 AND page_title = 'CATEGORYNAME'; [15:11:07] pmenon, ^ probably horrible inefficient but seems to work locally [15:11:13] horribly* [15:11:39] Krenair: how long did that query take ? [15:11:46] if it's hidden you'll get a row back [15:12:15] 0.00sec, but my test wiki DB is small, so... [15:12:38] It won't necessarily work so nicely on an enwiki dump :) [15:12:46] Krenair: hmm, I'll take a look [15:13:02] the id columns should be indexed, noe sure about the prop_name [15:13:05] and title [15:24:35] Krenair: why are you using page_namespace = 14 ? [15:25:23] That's the category namespace id on my wiki [15:25:54] And according to the JS 'wgNamespaceIds.category' on enwiki it's 14 there as well [15:27:44] gotcha [15:27:51] had a brain freeze moment [16:33:50] andre__: https://www.mediawiki.org/wiki/Project:Test_reports seems unused, should we just give up and redirect to [[How to report a bug]]? (It's also linked from http://deployment.wikimedia.beta.wmflabs.org which is used mainly by WMF staff anyway.) [16:42:30] Nemo_bis, didn't know about that page and looks like the last feedback item was added in March 2012 [16:42:37] yes [16:42:55] linking to [[How to report a bug]] could also be under "If you're comfortable entering bugs in Bugzilla" [16:43:03] so I have no idea - "be bold"? :) [16:43:24] * andre__ looks at https://www.mediawiki.org/wiki/Special:WhatLinksHere/Project:Test_reports [17:03:21] andre__: the page also had a twin on Meta but I killed that one [17:03:26] long time ago [17:03:29] heh [17:03:39] hello Raymond_ [17:50:20] ping [17:52:25] anomie: have a few minutes? there's a lua script we're using on wikidata that is going over the execution time and i was wondering if you could suggest any optimizations [17:52:35] legoktm: Sure [17:52:43] ok so the module is https://www.wikidata.org/wiki/Module:WBHacks [17:53:21] hmm ok at least there is chatter on this channel [17:53:21] and the page it is timing out on is https://www.wikidata.org/wiki/Wikidata:P [17:55:29] heh ping timeout [17:57:02] Just nod if you can hear me. [17:57:19] * legoktm nods [17:57:27] yay thank you [18:00:51] legoktm: Ugh. Your slowest things are fetching the content of those property pages and then parsing that content as JSON. [18:01:22] yeah…. [18:01:59] is there no native JSON module bundled with the extension? I didn't see one in the docs so I found that one online somewhere [18:05:08] No, there isn't. Bug 45470 requests one, but no one has come through with a compelling use case. [18:07:08] hm [18:07:22] Krenair: is there a bug for inconsistency in thread summary parsing? https://translatewiki.net/wiki/Thread:Support/About_MediaWiki:Translate-checks-balance/en [18:07:37] summary is parsed in talks and even RC, but not in thread NS [18:08:06] Well this module is really meant to be a temporary thing until we can use the native Wikibase Lua on the site, but I'm worried that that might still have the same issues [18:13:22] Native Wikibase Lua should be faster, I would think, since it wouldn't be converting to json and back. And I'd suspect the direct fetch of the info might be faster than going via title:getContent() too. [19:25:07] are enotifs slow for everyone now that jobqueue is higher? [19:38:29] Was there some delay/issue with jobqueue and mail? got several talkpage-messages oldest from 6:10 utc and mail-sendtime 16:31-16:32 in the last hours (ex. transferred to ISP at 19:20) [19:39:30] (this is for dewiki) [21:02:52] g-dang bz is annoying sometimes [21:03:13] Reedy: nevermind [21:03:17] :D [21:10:23] greg-g, I'd like to a lightning deploy today if possible. [21:10:25] It's https://gerrit.wikimedia.org/r/#/c/62479/ [21:10:43] Basically, small CSS and JS changes, and a PHP change to add the module. [21:11:20] superm401: sure, E2/Echo is using it right now, so check in with them. [21:12:07] greg-g, meaning they're doing a deploy now, or during the window (16:00 SF time)? [21:12:11] I was planning to use the window. [21:13:09] superm401: they were plannig on using a portio nof the LD (it was in parens on the calendar, I'm calling it out more now with this in-progress edit [21:13:12] ) [21:14:46] kaldari, do you think E3 can squeeze a small change into the window after your lightning deploy? [21:15:14] sure if there's any time left, but that's a big maybe :P [21:15:41] we're deploying a core change and a few Echo bugfixes [21:16:04] hopefully should go smoothly, but you never know [21:18:06] No problem [21:18:20] greg-g, can we reserve the lightning window for tomorrow? [21:18:53] superm401: sure [21:19:00] Thanks, I'll edit [21:19:42] oh, almost done [21:19:52] done [21:20:13] greg-g, okay, thanks. [21:50:35] gn8 folks [23:14:28] Hi folks. [23:17:02] I'm having trouble connecting to Wikipedia using HTTPS. I'm getting served by the Dutch cluster, specifically wikipedia-lb.esams.wikimedia.org right now (it was xe-1-1-0.cr2-knams.wikimedia.org earlier tonight). Is this a known problem? [23:17:41] idunn, please describe the problem. Do you know how to use curl? [23:17:49] More or less, yes. [23:18:35] Okay, do: [23:18:43] curl -I -v https://en.wikipedia.org [23:18:52] ldunn, you can change the language code if you want [23:18:54] LeslieCarr: esams trouble ----^^ [23:19:06] seriously ? [23:19:08] gah [23:19:13] also, can you traceroute [23:19:51] i had almost snuck out of the office ! [23:20:16] superm401: this is what I'm getting from curl: [23:20:16] * About to connect() to en.wikipedia.org port 443 (#0) [23:20:16] * Trying 91.198.174.225... [23:20:16] * 0x8001f2b0 is at send pipe head! [23:20:16] * STATE: CONNECT => WAITCONNECT handle 0x80057550; line 1032 (connection #0) [23:20:16] * Trying 2620:0:862:ed1a::1... [23:20:16] * Connected to en.wikipedia.org (2620:0:862:ed1a::1) port 443 (#0) [23:20:17] * error setting certificate verify locations: [23:20:17] CAfile: /usr/ssl/certs/ca-bundle.crt [23:20:18] CApath: none [23:20:18] * Closing connection 0 [23:20:19] * The cache now contains 0 members [23:21:17] Alright, that's weird. Hopefully LeslieCarr can take it from here. :) [23:21:40] OK, that may not have made it to the channel. Here's a copy on pastebin just in case: http://pastebin.com/aP6iaWZH [23:21:49] traceroute should hopefully help … also if i could have your ip , even if it's in a pm [23:22:06] LeslieCarr: mine? [23:22:13] yes please idunn [23:22:25] is it just https actually ? [23:22:34] Yeah, http is working. [23:22:49] hrm [23:22:49] ok [23:24:36] Traceroute shows a few local hops, then this: [23:24:36] 3 13 ms 11 ms 11 ms 145.254.10.201 [23:24:37] 4 33 ms 23 ms 23 ms 92.79.213.138 [23:24:37] 5 22 ms 21 ms 23 ms xe-1-1-0.cr2-knams.wikimedia.org [195.69.145.176] [23:24:37] 6 22 ms 23 ms 22 ms wikipedia-lb.esams.wikimedia.org [91.198.174.225] [23:27:21] i'm able to reverse traceroute to the 145.254.10.201 point [23:27:55] this is strange though, not seeing any depooling of the ssl cluster …. [23:28:56] Someone was reporting HTTPS issues in #wikipedia-en earlier... [23:29:08] Why can't I access wikipedia via https? [23:29:25] They're still logged into IRC [23:29:46] another person ? [23:29:58] can you get them to traceroute / curl as well ? [23:30:18] grr… so ssl3001-3003 appear to be functioning perfectly normally .... [23:31:54] I'm just looking at this in Wireshark. There's a bunch of SYNs being sent, but I'm not getting any reply whatsoever from wikipedia-lb.esams.wikimedia.org / 91.198.174.225. [23:32:06] mutante: http://pastebin.com/aP6iaWZH [23:43:32] ok i think i see what is happening though i don't know why [23:43:44] amslvs1 has the ip of 91.198.174.225 [23:44:04] nevermind, ipv6 [23:44:07] hrm [23:44:48] LeslieCarr, http://pastebin.com/Qwb9B0EU [23:45:24] thanks Krenair [23:46:25] idunn: are you having the same issues on https://en.wikinews.org ? [23:46:57] LeslieCarr: *checks* Yes. [23:47:09] what are the issues? [23:47:16] it's working on 80 and not 443? [23:47:21] Ryan_Lane: exactly. [23:48:04] and it doesn't appear to be just one -lb [23:48:14] but esams only [23:48:17] yep [23:48:27] can you not connect, or is it giving a cert error? [23:49:01] I can't connect at all. There's never a reply of any kind to the initial SYN packet. [23:49:42] one sec [23:50:37] so, from the LVS layer it's fine [23:50:40] it works for me through that IP [23:50:41] idunn: which country are you in? [23:50:46] Ryan_Lane: Germany. [23:52:21] hm [23:52:44] disregard that, I'm not testing it correctly [23:54:07] and he's got layer3 connectivity [23:54:11] ok, it is actually working for me [23:54:21] and we can traceroute back out from ssl3001-3003 to his gateway [23:55:01] so, if one of the ssl servers isn't working properly for ipv6 https, then it may affect some and not others [23:55:05] the SSL eqiad cluster shows a normal pattern of traffic [23:55:07] since it's using source hash [23:55:17] I mean esams [23:58:14] Does anyone know what happened to the optional clock and sandbox link in the user toolbar? They seem to have disappeared, and I don't see a bug filed on this issue when I search Bugzilla but maybe I'm doing it the wrong way. Can anyone help?