[01:30:27] <ToAruShiroiNeko>	 can there be an effort to translate catonese?
[01:30:33] <ToAruShiroiNeko>	 it tends to fail to be translated
[01:33:27] <Reedy>	 Interface language?
[01:33:41] <Reedy>	 Find people to do it and point them at translatewiki
[01:34:55] <ToAruShiroiNeko>	 its a lot of languages
[01:35:18] <Reedy>	 I didn't ask that
[01:35:19] <ToAruShiroiNeko>	 I wish to reach someone whom handles such stuff
[01:35:28] <ToAruShiroiNeko>	 whom can handle this for me
[01:35:30] <Reedy>	 Your best bet is Siebrand
[01:35:32] <Reedy>	 It's done by volunteeers
[01:35:37] <Reedy>	 You can't make people translate stuff
[01:35:47] <ToAruShiroiNeko>	 can you actually tell me the languages cattonese does not have a translation?
[01:36:00] <Reedy>	 Sorry, what?
[01:36:54] <ToAruShiroiNeko>	 zh-yue.
[01:36:58] <Reedy>	 http://translatewiki.net/w/i.php?title=Special:Translate&group=ext-0-all&language=yue
[01:36:59] <Reedy>	 20%
[01:37:36] <Reedy>	 83% for MW core
[01:38:33] <ToAruShiroiNeko>	 ah thats not what I was looking for
[01:38:50] <ToAruShiroiNeko>	 http://commons.wikimedia.org/wiki/Template:Assessments/translate/mt
[01:38:55] <ToAruShiroiNeko>	 check the last template
[01:39:11] <ToAruShiroiNeko>	  Ċiniż language then Cantonese language
[01:39:24] <ToAruShiroiNeko>	 Ċiniż is "Chineese" in mt language
[01:39:45] <ToAruShiroiNeko>	 no yue equavalent for mt
[01:40:10] <Reedy>	 So fix it?
[01:40:11] <Reedy>	 Add one
[01:40:15] <Reedy>	 You know how wikis work
[01:57:26] <ToAruShiroiNeko>	 Reedy the thing is I am trying to figure out what isnt working
[01:58:29] <Reedy>	 ..?
[01:59:14] <ToAruShiroiNeko>	 I believe the code is {{#language:Catonese|xx}}
[01:59:21] <ToAruShiroiNeko>	 *I believe the code is {{#language:Cantonese|xx}}
[01:59:47] <ToAruShiroiNeko>	 what am I trying to get treanslated on mediawiki?
[01:59:59] <ToAruShiroiNeko>	 its not regular interface probably
[02:01:17] <Reedy>	 https://www.mediawiki.org/wiki/Help:Magic_words#Miscellaneous
[02:01:24] <Reedy>	 it's a parser function
[02:01:39] <ToAruShiroiNeko>	 yes
[02:01:48] <ToAruShiroiNeko>	 but what needs to be translated here?
[02:03:14] <Reedy>	 $lang = Language::fetchLanguageName( $code, $inLanguage );
[02:04:37] <ToAruShiroiNeko>	 ok...
[02:04:53] <ToAruShiroiNeko>	 so $inLanguage holds the complete list?
[02:05:20] <Reedy>	 cldr
[02:05:20] <Reedy>	 $wgHooks['LanguageGetTranslatedLanguageNames'][] = 'LanguageNames::coreHook';
[02:05:46] <Reedy>	 It's an upstream thing
[02:05:46] <Reedy>	 http://cldr.unicode.org/
[02:06:20] <Reedy>	 If you look at CldrNamesMt, there is no 'yue' entry
[02:06:25] <Reedy>	 which is the problem here
[02:06:57] <Reedy>	 There's apparently also no Yue file, so you can't get any translates as they would be in cantonese
[02:07:19] <Reedy>	 http://cldr.unicode.org/index/bug-reports
[02:07:37] <Reedy>	 Stuff could be added to the cldr extension temporarily
[02:07:45] <Reedy>	 but it needs to go upstream, we shouldn't have to maintain it
[02:13:48] <ToAruShiroiNeko>	 ah I see
[02:14:03] <ToAruShiroiNeko>	 there are several ways to say cantonese though
[02:15:00] <ToAruShiroiNeko>	 seems like they use zh-yue instead of yue
[02:16:10] <ToAruShiroiNeko>	 or not
[02:16:14] <ToAruShiroiNeko>	 I am confused by this site
[02:54:39] <ToAruShiroiNeko>	 ***MEMORY-ERROR***: rsvg-convert[9385]: GSlice: failed to allocate 496 bytes (alignment: 512): Cannot allocate memory
[03:05:26] <TimStarling>	 ToAruShiroiNeko: make your SVGs smaller
[03:06:29] <ToAruShiroiNeko>	 not mine
[03:06:50] <ToAruShiroiNeko>	 its 23.85 MB in size
[13:07:50] <gerrit-wm>	 New patchset: Hashar; "<git> now supports 'failonerror' passed to <exec/>" [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/14477
[13:07:51] <gerrit-wm>	 New patchset: Hashar; "ant: <git-clone> supports options and failonerror" [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/14478
[13:07:51] <gerrit-wm>	 New patchset: Hashar; "ant: redo copy-project to use --ref cloning" [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/14479
[13:08:32] <gerrit-wm>	 Change merged: Hashar; [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/14479
[13:08:33] <gerrit-wm>	 Change merged: Hashar; [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/14478
[13:08:34] <gerrit-wm>	 Change merged: Hashar; [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/14477
[13:26:50] <hashar>	 13:26:06       [exec] error: RPC failed; result=22, HTTP code = 502
[13:26:51] <hashar>	 ...
[14:58:42] <gerrit-wm>	 New patchset: Hashar; "ant: target to easily apply a change" [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/14486
[14:58:43] <gerrit-wm>	 New patchset: Hashar; "ant: experimental Wikidata build target" [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/14487
[14:59:15] <gerrit-wm>	 Change merged: Hashar; [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/14487
[14:59:16] <gerrit-wm>	 Change merged: Hashar; [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/14486
[15:21:10] <gerrit-wm>	 New patchset: Hashar; "ant: apply change macro get refspec/branch params" [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/14489
[15:21:10] <gerrit-wm>	 New patchset: Hashar; "ant: Wikidata build did not correctly apply change" [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/14490
[15:21:30] <gerrit-wm>	 Change merged: Hashar; [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/14489
[15:21:43] <gerrit-wm>	 Change merged: Hashar; [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/14490
[15:57:23] <gerrit-wm>	 New patchset: Hashar; "ant: make sure we import a clean & uptodate Gerrit project" [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/14496
[15:58:31] <gerrit-wm>	 Change merged: Hashar; [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/14496
[16:06:20] <hashar>	 hmm
[16:06:33] <hashar>	 almost able to build test out Wikidata using Jenkins!
[16:10:22] <chrismcmahon>	 hashar: what do you think about deploying to beta labs (including some extensions) with Jenkins?
[16:10:47] <hashar>	 hey chris :)
[16:11:15] <hashar>	 chrismcmahon: I have no idea :( still a goal I want to achieve though
[16:11:16] <chrismcmahon>	 hey Antoine :)  nice work on Wikidata and Jenkins
[16:11:24] <hashar>	 would need a funny script to handle that
[16:11:39] <hashar>	 the Wikidata I have been hacking it for most of the week :/
[16:11:55] <chrismcmahon>	 hashar: I was reading about deploying PHP to Ubuntu with Jenkins yesterday, there are some good articles via Google
[16:12:58] <hashar>	 chrismcmahon: I did wrote some doc at https://labsconsole.wikimedia.org/wiki/Deployment/Overview
[16:13:22] <hashar>	 still need to write some maintenance doc such as how to update extensions, how to update core, update db ...
[16:15:41] <hashar>	 chrismcmahon: I did a demo of Jenkins yesterday at my local hacker group
[16:15:51] <hashar>	 (well more a group where we troll and drink beer, but still)
[16:15:56] <chrismcmahon>	 ;0
[16:16:14] <hashar>	 some expressed interest and did not even knew about that software
[16:16:30] <hashar>	 most probably because it is a Java world tool and the group does not play Java :-D
[16:17:44] <chrismcmahon>	 hashar: it's way beyond Java, it's pretty much the industry standard CI tool anymore, much more popular than CruiseControl I think these days
[16:19:08] <hashar>	 probably
[16:21:27] <chrismcmahon>	 anyway, it would be great to see Mediawiki and the AFT/E2 projects deployed to beta labs from Jenkins.  from your doc, it sounds like that might be close.
[16:21:28] <^demon>	 chrismcmahon: Indeed, we played with cruise control before using jenkins.
[16:21:56] <^demon>	 cruise control was a pita to configure and didn't give nearly as useful output
[16:23:32] <chrismcmahon>	 yeah, I don't think Thoughtworks wants to put in the effort to really have CC compete with Jenkins. CC broke the ground, but Jenkins is pretty much the future from what I can tell.
[16:25:12] <gerrit-wm>	 New patchset: Hashar; "(bug 37050) enable Jenkins for Wikidata" [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/14500
[16:25:41] <gerrit-wm>	 Change merged: Hashar; [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/14500
[16:26:36] <hashar>	 enough for this afternoon, will be back later this evening
[20:45:22] <Platonides>	 how big is ARticleFeedbackv5 for you?
[20:48:56] <Platonides>	 hmm CategoryTree is 6.5M locally but 1.5M after a new clone
[20:49:04] <Platonides>	 and 7,4M after a repack!
[20:50:18] <^demon>	 How did you repack it?
[20:50:28] <Platonides>	 git repack -A -F
[20:50:54] <Platonides>	 and then again with --window=20 hiping it would improve
[20:51:04] <^demon>	 No -d?
[20:51:13] <^demon>	 -d removes redundant packs.
[20:52:15] <Platonides>	 I just looked up the incantation :P
[20:52:31] <^demon>	 My usual repack is `git repack -a -d -f --depth=250 --window=250`
[20:52:41] <^demon>	 Which is roughly what you'd get with `git gc --aggressive`
[20:53:04] <Platonides>	 how big is your ArticleFeedbackv5 repo?
[20:53:16] <^demon>	 I don't think I have it cloned, one sec.
[20:54:03] <^demon>	 6.2 on fresh clone.
[20:54:17] <Platonides>	 75M here :P
[20:54:56] <Platonides>	 before -> 75M
[20:54:58] <Platonides>	 git repack -A -F -> 78M
[20:55:05] <Platonides>	 git repack -A -F -d -> 6,4M
[20:55:16] <Platonides>	 git repack -a -d -f --depth=250 --window=250 -> 5. M
[20:55:23] <Platonides>	 *5.0M
[20:55:28] <^demon>	 Fresh clone -> 6.2M.
[20:55:38] <^demon>	 git repack -A -F -> 5.9M
[20:56:08] <^demon>	 git repack -A -F -d -> 6.0M
[20:56:25] <Platonides>	 I think I'll start repacking repos
[20:57:09] <^demon>	 Generally git is pretty good about garbage collection, but usually doing some repacking helps.
[20:57:16] <Platonides>	 one wonders why should that be done manually...
[20:57:17] <^demon>	 Core especially has the habit of getting absurdly large.
[20:58:16] <Platonides>	 well, in this case it shouldn't need to garbage-collect
[20:58:23] <Platonides>	 i don't think I have committed anything there
[20:58:23] <^demon>	 git repack -a -d -f --depth=250 --window=250 -> 4.8M
[20:58:24] <^demon>	 :)
[20:58:47] <Platonides>	 different git version?
[20:59:09] <^demon>	 git version 1.7.7.5 (Apple Git-26)
[20:59:18] <Platonides>	 git version 1.7.11.1
[20:59:54] <Platonides>	 maybe the blobs were considered in different order...
[20:59:58] <^demon>	 I wonder if delta compression is OS/platform-specific in some manner.
[21:00:12] <Platonides>	 I don't see how it would be
[21:00:24] <Platonides>	 also, isn't that little-endian, too?
[21:00:52] <Platonides>	 well, maybe Linus put some arch-specific optimizations
[21:02:33] <^demon>	 yeah, osx is little endian. i honestly don't know really.
[21:03:02] <Platonides>	 how many cores?
[21:03:15] <^demon>	 Dual.
[21:03:28] <Platonides>	 so no difference there, either
[21:05:58] <Platonides>	 6,2M	ArticleFeedbackv5-cloned
[21:05:58] <Platonides>	 4,8M	ArticleFeedbackv5-cloned-repacked
[21:06:12] <Platonides>	 so it was how it considered the blobs
[21:17:28] <Platonides>	 AbuseFilter: 73.1M
[21:17:47] <Platonides>	 sending your spell
[21:18:08] <^demon>	 I have that aliased as `git repack-everything` :)
[21:18:38] <Platonides>	 then I'll probably copy your alias name :)
[21:18:50] <Platonides>	 Translate: 76M
[21:19:19] <^demon>	 I really don't understand it sometimes :p
[21:19:47] <Platonides>	 what's "it"?
[21:20:19] <^demon>	 git ;-)
[21:24:06] <Platonides>	 heh :)
[21:24:32] <Platonides>	 it wasn't a good idea to start both repack-everything at the same time
[21:24:45] <Platonides>	 it started swapping :P
[21:25:18] <Platonides>	 not really strange if one of them uses 2GB of memory...
[21:26:55] <Platonides>	 why isn't the total of "Compressing objects" the same as the output of counting objects?
[21:30:35] <Platonides>	 AbuseFilter went down to 7.4M
[21:30:39] <Platonides>	 almost 10%
[21:31:43] <^demon>	 Maybe not all objects could be compressed?
[21:31:45] <^demon>	 Dunno
[21:35:51] <Platonides>	 Trnaslate 76M -> 15M
[21:36:00] <Platonides>	 ^demon?
[21:36:24] <Platonides>	 73.1M -> 7.4M is a very good compression
[21:36:58] <Platonides>	 WebFonts is also big, but this contains big objects
[23:08:24] <ToAruShiroiNeko>	 jorm are you around by any chance?
[23:08:47] <ToAruShiroiNeko>	 could you tell me the number of edits commons can take every minute?
[23:09:13] <jorm>	 i am.
[23:09:23] <ToAruShiroiNeko>	 yay
[23:09:37] <jorm>	 no; i suggest you ask. . . tim starling.
[23:09:45] <ToAruShiroiNeko>	 oh ok
[23:09:49] <ToAruShiroiNeko>	 disregard the pm then
[23:09:57] <jorm>	 since he had a thing about this recently, iirc.
[23:10:01] <jorm>	 or disabled someone's bot.
[23:10:06] <ToAruShiroiNeko>	 yes
[23:10:23] <jorm>	 performance statistics are not my forte.
[23:10:39] <jorm>	 roan might know as well.  also aaron shultz.
[23:10:42] <jorm>	 maybe reedy.
[23:32:05] <Platonides>	 UploadWizard was 64M -> 6.7M