[03:23:12] <gerrit-wm>	 New patchset: Krinkle; "(bug 34815) Use QUnit ?module= query parameter." [integration/testswarm] (master) - https://gerrit.wikimedia.org/r/11830
[08:05:49] <gerrit-wm>	 New review: Hashar; "(no comment)" [integration/testswarm] (master); V: 1 C: 2;  - https://gerrit.wikimedia.org/r/11823
[08:05:51] <gerrit-wm>	 Change merged: Hashar; [integration/testswarm] (master) - https://gerrit.wikimedia.org/r/11823
[08:10:37] <gerrit-wm>	 New patchset: Hashar; "Update path to testswarm submit script." [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/11824
[08:12:43] <gerrit-wm>	 New review: Hashar; "Patch forgot to update the testswarm submodule :-D" [integration/jenkins] (master); V: 1 C: 2;  - https://gerrit.wikimedia.org/r/11824
[08:12:45] <gerrit-wm>	 Change merged: Hashar; [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/11824
[08:25:23] <gerrit-wm>	 New review: Hashar; "I manually triggered a job run for https://gerrit.wikimedia.org/r/#/c/9504/" [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/11824
[08:32:39] <gerrit-wm>	 New review: Hashar; "(no comment)" [integration/testswarm] (master); V: 0 C: 0;  - https://gerrit.wikimedia.org/r/11830
[09:09:23] <ToAruShiroiNeko>	 I get a Unknown error: "internal_api_error_UploadChunkFileException".
[09:09:45] <ToAruShiroiNeko>	 when I try to upload a 347MB
[09:09:49] <ToAruShiroiNeko>	 on commons
[11:29:39] <Daniel_WMDE>	 uh, the phpstorm license is expiring TODAY. do we have a now one?
[11:29:43] <Daniel_WMDE>	 who is looking after that?
[12:30:03] * werdna  waves

[13:41:19] <apergos>	 does anyone know how in gerrit someone can enable sparse checkout on their repo (ui, command line, whatever)?
[13:41:27] <apergos>	 or have we decided this is not ok to do?
[13:44:51] <hashar>	 I don't even know what spare checkout is
[13:44:53] <hashar>	 :/
[13:45:35] <hashar>	 "Sparse checkout" allows populating the working directory sparsely.
[13:45:39] <hashar>	 is not going to help from git doc
[13:45:55] <hashar>	 something like:   Azzzzzzz stuff is  doing stuff Azzzzzzzzly
[13:45:56] <hashar>	 ;)
[14:58:13] <Nemo_bis>	 Platonides, ping
[16:23:14] <sumanah>	 hi there chrismcmahon
[16:23:25] <chrismcmahon>	 hi sumanah
[16:23:34] <sumanah>	 chrismcmahon: how is the depacking going? :-)
[16:24:32] <chrismcmahon>	 it sort of looks like someone lives here :)  I'm going to actually cook dinner for the first time in weeks (as soon as I buy some groceries)
[16:24:55] <chrismcmahon>	 sumanah: I'm cranking on that blog entry we discussed, I should have it done today
[16:25:16] <sumanah>	 chrismcmahon: oh sweet! I hope it is useful.
[16:25:36] <sumanah>	 chrismcmahon: btw you may find this interesting http://bethgsoc.blogspot.com/2012/06/music21-version-10-released.html
[16:25:48] <sumanah>	 "Music21 is a python package designed to help scholars, researchers, and all music lovers answer computationally intensive questions about music quickly and simply. This python module provides extensive methods to analyze symbolic music, extracting information that has been hidden from listeners for centuries. It interfaces with a variety of formats, from musicxml to vexflow to even braille music output."
[16:25:51] <Nikerabbit>	 didn't labs have irc channel?
[16:26:20] <sumanah>	 yes it does
[16:26:25] <sumanah>	 #wikimedia-labs I think Nikerabbit
[16:27:49] <Nikerabbit>	 thanks
[16:28:00] <Nikerabbit>	 sumanah: that channel is not mentioned anywhere in labsconsole wiki :(
[16:28:14] <sumanah>	 Nikerabbit: go ahead & mention it.
[16:28:24] <sumanah>	 I think it was, & must have been deleted
[16:28:44] <Nikerabbit>	 ah but there it is right on my face... just not where i was looking for it
[16:28:53] <sumanah>	 go ahead & put it where you were looking for it
[16:29:03] <sumanah>	 Nikerabbit: have you ever heard of the landscape architecture principle "desire paths"?
[16:29:12] <chrismcmahon>	 that's pretty slick sumanah.  I've said any number of times that people wail about the decline in music education, but I think that between software like that and the amazing range of music available on youtube etc.,  another few years are going to see an explosion of great young musicians
[16:29:15] <sumanah>	 wherever you were looking for it, that's someplace that should have a pointer to the right thing
[19:35:16] <sumanah>	 chrismcmahon: http://wikimedia.de/wiki/English_Version in case you know QA people who should work on Wikidata
[19:37:56] <chrismcmahon>	 thanks sumanah I'll think about that one
[19:57:49] <gerrit-wm>	 New patchset: Hashar; "`Gerrit-Set-Reviewers` job to add reviewers" [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/11939
[19:58:04] <gerrit-wm>	 New review: Hashar; "(no comment)" [integration/jenkins] (master); V: 1 C: 2;  - https://gerrit.wikimedia.org/r/11939
[19:58:06] <gerrit-wm>	 Change merged: Hashar; [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/11939
[19:59:51] <gerrit-wm>	 New patchset: Hashar; "really ignore `workspace` directories" [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/11940
[20:00:06] <gerrit-wm>	 New review: Hashar; "(no comment)" [integration/jenkins] (master); V: 1 C: 2;  - https://gerrit.wikimedia.org/r/11940
[20:00:08] <gerrit-wm>	 Change merged: Hashar; [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/11940
[21:32:54] <hashar>	 Krinkle: I did merge your change to testswarm submitted
[21:33:01] <hashar>	 Krinkle: tested one, seems to work fine :-]
[21:33:06] <Krinkle>	 great
[21:33:16] <hashar>	 Krinkle: if you find any issue, please drop a bold email :-]
[21:33:29] <hashar>	 though I am heading bed right now ...
[21:33:34] <hashar>	 thanks for the patch!
[21:33:39] <Krinkle>	 hashar: one minute
[21:33:46] <Krinkle>	 hashar: So I want to get going with browserstack and all. I'm blocked on not being able to do stuff.
[21:34:17] <hashar>	 ohhh
[21:34:19] <Krinkle>	 I have already tested most things locally and with jquery, its just a matter of implementation. But as we all know, that can be simple or hard, depends, we'll see how it goes. But I do like to get started on this.
[21:34:21] <hashar>	 what blocks you?
[21:34:53] <Krinkle>	 hashar: 1) Not having jenkins in labs (whether or not temporarily for testing only), 2) Knowing nothing about puppet
[21:35:09] <hashar>	 puppet, you will learn
[21:35:15] <hashar>	 or you will get someone to do the job for you
[21:35:17] <Krinkle>	 I'd like you to help me get Jenkins up and running one way or another, then I can get stuff implemented and puppetize later.
[21:35:20] <hashar>	 so it is a non issue to me
[21:35:26] <hashar>	 the first is actually a blocker
[21:35:33] <Krinkle>	 yep
[21:35:40] <hashar>	 the Jenkins puppet class are not able to install jenkins as I understand it
[21:35:52] <hashar>	 maybe we should create a continuous integration project in labs ?
[21:35:57] <Krinkle>	 Yes
[21:36:03] <hashar>	 and have an instance for gerrit + jenkins + one for testswarm ?
[21:36:20] <Krinkle>	 Oh, right. we need a virtual gerrit as well
[21:36:21] <Krinkle>	 Yes, lets do that
[21:36:34] <Krinkle>	 hashar: wait, don't jerkins and testswarm need to be on the same instate ?
[21:36:36] <hashar>	 would you mind opening related bugs ? :-]
[21:36:36] <Krinkle>	 instance*
[21:36:38] <Krinkle>	 they share files
[21:36:55] <Krinkle>	 in a way, it is copied from one to the other.
[21:37:00] <hashar>	 yeah jenkins / testswarm need to be on the same server for now
[21:37:03] <Krinkle>	 ok
[21:37:08] <hashar>	 but we could as well use shared filesystem
[21:37:16] <hashar>	 labs as /data/project shared among all instances of a project
[21:37:24] <Krinkle>	 tried that before, I'd prefer not to wait 20 seconds per http request
[21:37:32] <Krinkle>	 for now :)
[21:37:34] <hashar>	 ;)
[21:37:49] <hashar>	 I guess first step would be to get a continuous integration project in labs
[21:38:07] <hashar>	 if you can open a bug for that and ask for the creation on labs … that would be a good first step
[21:38:13] <Krinkle>	 ok
[21:38:18] <Krinkle>	 I'll do that now, should be done by tomorrow
[21:38:28] <hashar>	 meantime, create a virtual instance in some project you have access to
[21:38:42] <hashar>	 and try installing jenkins using the existing class
[21:38:49] <hashar>	 should be misc::content::jenkins or something
[21:38:57] <Krinkle>	 I wouldn't have a clue on where to start doing that
[21:39:06] <hashar>	 that will give you ton of error message, from there we can start enhancing the puppet classes :-]
[21:39:11] <Krinkle>	 ok
[21:39:48] <hashar>	 given your work hours, I guess you will want to find someone from ops to assist you
[21:40:06] <hashar>	 maybe schedule a training session with Roan ?
[21:41:02] * RoanKattouw  runs away

[21:41:02] <Krinkle>	 OK
[21:41:13] <Krinkle>	 hashar: who should I assign the labs bug to?
[21:41:20] <hashar>	 Krinkle: the puppet class is misc::contint::test
[21:41:24] <Krinkle>	 OK
[21:41:33] <hashar>	 then we have several subclasses such as misc::contint::test::jenkins
[21:41:50] <hashar>	 I think projects are created by ops
[21:42:00] <hashar>	 I know Ryan and Daniel Zahn can do them
[21:42:04] <hashar>	 maybe someone from SF team
[21:42:08] <hashar>	 make sure to join #wikimedia-labs
[21:42:32] <hashar>	 maybe someone else can help there :)
[21:42:46] <hashar>	 anyway bed time
[21:43:00] <hashar>	 feel free to email me with whatever come to your mind and cause you any issue :-]
[21:43:09] <hashar>	 if I can't solve them, I will find you someone to assist hehe
[21:43:18] <hashar>	 meanwhile sleep time sorry
[21:43:41] <Krinkle>	 sure, thanks for the help so far :)
[21:43:56] <Krinkle>	 RoanKattouw: ::rolleyes::
[21:44:10] <RoanKattouw>	 heh
[21:44:17] <RoanKattouw>	 I'm busy! :D
[21:44:32] <hashar>	 Krinkle: by the way there is a Jenkins project in labs!
[21:44:46] <Krinkle>	 *another* one?
[21:44:59] <Krinkle>	 projects still can't be deleted?
[21:45:13] <Krinkle>	 hashar: btw, if you want to do, please do.
[21:45:15] <Krinkle>	 go*
[21:45:18] <hashar>	 I made you a sysadmin + netadmin to the Jenkins project
[21:45:23] <hashar>	 so you can create an instance there now :-]
[21:45:26] <Krinkle>	 ok
[21:45:38] <hashar>	 https://labsconsole.wikimedia.org/wiki/Nova_Resource:Jenkins
[21:45:56] <hashar>	 looks like someone named Catrope worked on that at some point
[21:46:09] <hashar>	 I would just create another instance and try the class above
[21:46:10] <Krinkle>	 I love how apple's spelling correction corrects jenkin -> jerkin, and then I type the 's'
[21:46:17] <hashar>	 then note the error and try to find out how puppet works :-]
[21:46:33] <RoanKattouw>	 Yeah I set it up back in the day and then abandoned it
[21:46:47] <Krinkle>	 hashar: Can I test puppet changes without merging? (e.g. do I need to set up a local puppet master?)
[21:47:00] <hashar>	 not that I know of :-(
[21:47:08] <hashar>	 need to be merged in test first
[21:47:17] <Krinkle>	 oh, we have a separate branch, okay
[21:47:17] <hashar>	 then wait a minute for a cronjob to apply the change to the labs puppet master
[21:47:24] <Krinkle>	 still annoying though
[21:47:25] <hashar>	 then update manually using:  puppetd -tv
[21:47:36] <hashar>	 paravoid (Faidon) is working on a new way
[21:47:38] <hashar>	 based on branches
[21:47:50] <hashar>	 so people could submit/merge their changes in their branches without waiting for ops
[21:48:00] <hashar>	 then people will have the responsibility to upstream their changes to the production branch
[21:48:07] <Krinkle>	 still, having to go through gerrit for a test is a broken workflow. I'd be making 20 commits of which 19 break puppet probably until I get it right in the beginning.
[21:48:11] <hashar>	 hence scaling out the review process hopefully
[21:48:23] <hashar>	 yeah
[21:48:28] <hashar>	 that is steep learning curve
[21:48:32] <Krinkle>	 Just being able to change a file on the local instance and e.g. pipe it to puppet or somethjing
[21:48:37] <hashar>	 you can also install puppet locally
[21:48:58] <hashar>	 through ruby gem IIRC
[21:48:59] <Krinkle>	 ah
[21:49:00] <Krinkle>	 maybe that should be the default, is it hard to setup?
[21:49:04] <Krinkle>	 k
[21:49:06] <Krinkle>	 I'll see
[21:49:15] <hashar>	 then to validate a file:   puppet parser validate manifests/some/file.pp
[21:49:19] <Krinkle>	 can you give me merge rights to test?
[21:49:20] <hashar>	 that at least let you parse the file
[21:49:27] <hashar>	 I don't have any rights on test
[21:49:28] <hashar>	 only ops do
[21:49:45] <hashar>	 access to puppet == root access
[21:49:48] * Krinkle  explodes

[21:49:50] <hashar>	 (somehow)
[21:49:52] <hashar>	 sorry ;-D
[21:49:59] <Krinkle>	 this is a broken workflow
[21:50:04] <hashar>	 yup
[21:50:09] <hashar>	 you can still play with jenkins in labs
[21:50:11] <Krinkle>	 I understand it though, but it isn't set up for labs as labs is.
[21:50:14] <Krinkle>	 yeah
[21:50:20] <hashar>	 I am off now
[21:50:23] <Krinkle>	 go!
[21:50:26] <hashar>	 let me know by email your finding if you want
[21:50:33] * hashar  goes to sleep

[21:50:37] <hashar>	 :-]
[21:50:44] <Krinkle>	 m=/hashar*go/g
[21:50:47] <Krinkle>	 m.length = 5
[21:51:17] <Krinkle>	 love that QUIT msg
[21:52:16] <Krinkle>	 RoanKattouw: I missed some of your RL2 commits, I know why not. They didn't' have me on it as reviewer.
[21:52:18] <Krinkle>	 adding myself now
[21:52:30] <Krinkle>	 now*
[21:53:17] <RoanKattouw>	 You can subscribe to extensions you know
[21:53:42] <Krinkle>	 oh?
[21:53:52] <Krinkle>	 but still, I want them on my dashboard.
[21:53:56] <RoanKattouw>	 You will have that too
[21:53:59] <RoanKattouw>	 You can watch extensions
[21:54:02] <Krinkle>	 ok
[21:54:24] <RoanKattouw>	 Go to Settings -> Watched Projects
[21:54:26] <Krinkle>	 hm.. there is no link anywhere to the Project page, I have to look it up in the registry.
[21:54:30] <Krinkle>	 oh under settings
[21:54:36] <RoanKattouw>	 And then watch mediawiki/extensions/Gadgets
[21:54:54] <RoanKattouw>	 You can choose to only watch the RL2 branch but it's probably better to just watch the entire repo, that's what I do
[21:55:17] <Krinkle>	 yeah
[21:55:29] <Krinkle>	 it says "Email Notifications" though
[21:55:33] <Krinkle>	 it adds them to my dashboard?
[21:55:41] <RoanKattouw>	 It does, yeah
[21:55:57] <RoanKattouw>	 Or.... wait
[21:56:00] <RoanKattouw>	 Maybe it doesn't?
[21:56:20] <RoanKattouw>	 Aha, there's a separate "Watched Changes" dashboard
[21:56:38] <Krinkle>	 yeah "Review Requests for Krinkle" is where I want them
[21:57:13] <RoanKattouw>	 Unfortunately that's not where they go
[21:57:26] <Krinkle>	 I can add them now, now that I'm notified
[21:57:39] <Krinkle>	 !g Icde35055
[21:57:39] <wm-bot>	 https://gerrit.wikimedia.org/r/#q,Icde35055,n,z
[21:57:56] <Krinkle>	 what's with that one? Should it be abandoned ? Or is it a todo?
[21:58:32] <RoanKattouw>	 It's a todo yeah
[21:58:47] <RoanKattouw>	 Although it's kind of dormant right now
[21:59:20] <Krinkle>	 gadget prefs is postponed for RL 2.x or 3.x
[21:59:41] <RoanKattouw>	 Yeah
[21:59:49] <RoanKattouw>	 Speaking of which
[22:00:00] <RoanKattouw>	 Have you made any progress with RL2 recently? Will you have time for that soonish
[22:00:03] <RoanKattouw>	 ?
[22:00:18] <RoanKattouw>	 If nothing else I was thinking we could do a sprint in the first week of July
[22:00:40] <RoanKattouw>	 (Sans the Wednesday of course, I have the 4th of July off)
[22:02:32] <Krinkle>	 I'm taking today to catch up on code review, I am planning to focus the rest of the week on getting your commits reviewed and a good chunk of the todo items.
[22:02:46] <Krinkle>	 (todays cr is mostly mw-core)
[22:04:05] <RoanKattouw>	 OK
[22:04:19] <RoanKattouw>	 Sounds good, let me know if you need anything
[22:04:42] <Krinkle>	 will do
[22:04:51] <RoanKattouw>	 I don't actually have time for RL2 myself, unfortunately, because of the VisualEditor release on Thursday
[22:05:22] <Krinkle>	 RoanKattouw: I was going to use wmf labs/rl2/repo1 as resource loader source, but it was too slow to use locally. Using prototype.wikimedia.org now as api source (will set up another wiki locally later).
[22:05:28] <Krinkle>	 Found out a nice bug straight away
[22:05:42] <RoanKattouw>	 Oh?
[22:05:47] <Krinkle>	 getRepoData (also used for foreign wikis) is forcing 'json' without callback.
[22:05:51] <RoanKattouw>	 Ah!
[22:05:58] <Krinkle>	 cross-orgin exception all over
[22:06:00] <RoanKattouw>	 Yeah
[22:06:03] <RoanKattouw>	 Needs jsonp
[22:06:07] <Krinkle>	 yep
[22:06:22] <Krinkle>	 I always thought jquery detected cross-domain internally and switched
[22:06:24] <Krinkle>	 it used to do that
[22:06:27] <Krinkle>	 I'll dig into that
[22:06:38] <Krinkle>	 it does do that for $.getJSON
[22:06:42] <Krinkle>	 we're using $.ajax
[22:06:47] <RoanKattouw>	 Perhaps if you set the type to jsonp?
[22:07:24] <Krinkle>	 Well, then we always use jsonp. That works, but would be nice to use ajax when possible instead of <script> tags
[22:07:41] <RoanKattouw>	 Yeah, true
[22:07:44] <Krinkle>	 although it is somewhat unreal for cross-wiki to be on the same domain
[22:08:00] <Krinkle>	 anyway, should be a simple solution. I know jquery is capable of handling it
[22:08:03] <RoanKattouw>	 Yeah, I was gonna say the same thing
[22:08:18] <RoanKattouw>	 We probably don't need to worry about it too much, two wikis on the same domain is uncommon
[22:09:14] <Krinkle>	 btw, I was hoping maybe one day we can use ajax to retrieve bits.wikimedia.org load.php, not sure about performance yet, but sounds nice (certainly saves dom-operations and better callbacks)
[22:09:38] <Krinkle>	 globalEval now supports setting a url in most browsers
[22:09:44] <Krinkle>	 jQuery uses that as of 1.8pre
[22:09:52] <Krinkle>	 so we get proper debugging still
[22:10:13] <RoanKattouw>	 What is globalEval?
[22:10:15] <Krinkle>	 which was the main reason we went with mw.html instead of $.getScript
[22:10:29] <Krinkle>	 $.globalEval, used by jQuery to evaluate ajax text/javascript
[22:10:33] <RoanKattouw>	 Oh OK
[22:10:35] <RoanKattouw>	 Cool
[22:10:55] <RoanKattouw>	 We'd need CORS for that though, wouldn't we?
[22:11:16] <Krinkle>	 $.globalEval is just a 1 liner to window.eval(src, data); where data is that I mentioned above
[22:11:19] <Krinkle>	 RoanKattouw: Yes
[22:11:28] <RoanKattouw>	 Right
[22:12:41] <Krinkle>	 basically  it is "DOM operation (inserting <script>) and 2 event listeners and clean up - with native evaluation" vs. "ajax with globalEval"
[22:13:05] <Krinkle>	 the former will always be the fallback, but jquery takes care of that