=== kadams54 is now known as kadams54-away === kadams54-away is now known as kadams54 === kadams54 is now known as kadams54-away === kadams54 is now known as kadams54-away === kadams54 is now known as kadams54-away === kadams54-away is now known as kadams54 === kadams54 is now known as kadams54-away === kadams54 is now known as kadams54-away [15:00] hatch: just open a MP against trunk, it'll make it's way into the queue [15:01] lazyPower: yup found that out :) took about a day to get into the queue :) [15:03] ahhh -30C gota love it [15:04] -42C with windchill [15:04] hatch: awesome! [15:04] :D [15:05] lol [15:06] it's really early for temps this low - I hope this isn't a sign of things to come [15:27] oh, man, hatch, i missed to comment the PR666! [15:28] lol you did [15:34] man this would be nice to have https://support.leankit.com/entries/30469547-Merge-2-Cards-Together [15:40] teslanick: hey I was looking at lisps this last weekend but could not find an answer to whether CL has any features that Clojure does not [15:41] CL has better macro support. [15:41] which isn't relaly too interesting because macros are only executed at compile [15:41] really* [15:42] That's actually most of the power of Lisp [15:42] yeah I know lol [15:43] in my head it's only syntactic sugar over just creating a wrapper function [15:43] maybe I'm thinking of that wrong [15:43] Yeah, definitely thinking about that wrong [15:44] Macros let you do really powerful source code transformations, which is also their biggest danger. It's easy to write a macro that makes your code incomprehensible to the average dev. [15:44] e.g. I wrote a Clojure macro that looks for symbols in the form and replaces them with instances of a Hessian remote interface. [15:44] do you know of any examples which show it being useful beyond just a wrapper fn? [15:45] it's essentially like a built in transpiler then? [15:46] In clojure, go-style CSP is implemented as a macro. [15:46] So, inside a (go ...) s-expression, the semantics of the language are entirely different. [15:49] blah I need to do more research - I was initially looking at Lisps for the macro support but couldn't find anything which 'clicked' to show the power other than being like a wrapper/transpiler [15:50] Well, it is a transpiler in a sense. It translates one set of symbols to a different set of symbols. For domain-specific problems it's a really valuable affordance. [15:50] I was actually initially looking for runtime macro support :) [15:51] but to do that you need a special version of CL [15:51] I don't know what you mean by "runtime macro" [15:51] In Clojure, the difference between read-time and run-time is pretty blurry. [15:52] e.g., I can write a macro in the repl and then immediately use it on my clojure code. [15:52] at compile Fn A calls fn B but depending on some interaction Fn A now calls G [15:52] uiteam call in 9 kanban please [15:52] and that's a source code change, not just a conditional [15:53] Well, you could write a macro that captures that specific logic. But to capture any arbitrary logic, you're folding a ton of logic into your build step [15:53] exactly [15:54] so there is a version of CL which maintains the source which you can then modify and recompile [15:55] but the CL landscape is a wash with custom versions heh [15:55] not sure I want to wade into that, which is why I started looking at Clojure - but that didn't quite get me where I wanted [15:56] js and it's prototype tree could be used to do something similar [15:56] well in that it would call the same fn name but it would be a different fn [15:58] Yeah, the problem with CL (as I understand it, I've not played with it at all, really) is that the moment you've bootstrapped yourself into good and usable code, you're no longer writing CL anymore. [15:58] Clojure is nice because it's way closer to good and usable code without a ton of bootstrapping. [15:59] CL also doesn't have a good entry point [15:59] But it's a tradeoff -- you don't get fully-featured reader macros. [16:00] Also, it really sounds like you want to use a clojure multimethod for the problem you have in mind [16:01] Which is basically a multiple-dispatch function based on some arbitrary logic. === kadams54 is now known as kadams54-away === kadams54-away is now known as kadams54 === kadams54 is now known as kadams54-away [16:07] I'll look into that [16:19] maybe I can come up with a way of doing it with goroutines [16:19] What problem are you trying to solve? [16:20] a totally arbitrary academic made up one :) [16:20] basically I wanted to investigate a script in which a function would be multi purposed [16:20] but not by conditionals but by code-rewriting [16:25] Well, from a technical standpoint, compiled clojure code can't do dynamic code rewriting because it all gets flattened into java bytecode. [16:25] Even in JS, you'd have to either write a meta-interpreter or eval() the rewritten code. [16:27] yeah I was hoping to find a language which allowed such a thing [16:28] I was even thinking something compiled (Go C C++) which could compile a new 'dll'esk module [16:32] Right, and you can do dynamic compilation in most languages (it's a pain in the ass), but it's a serious code smell. [16:32] I'm pretty sure you could write a clojure interpreter in clojure pretty easily. It comes with an excellent set of reader libs. [16:41] this is probably all nonsense anyways - but I couldn't find any papers or anything that actually showed it being functional [16:41] lots of talk about it [16:41] but no examples [16:41] but maybe my google fu is failing me === kadams54 is now known as kadams54-away === kadams54-away is now known as kadams54 [17:50] rick_h_: will there still be the category drop down with the ac removal? [17:51] hatch__: no, because we'll be moving to tags and a different search/filter method in the future [17:52] sounds like a plan [17:52] *ctrl+a del* [17:52] :) [18:25] I can tell Huw wrote the code I'm working on right now because 'else' is on a new line [18:26] lol [18:26] #idotoomanycodereviews [18:27] :P === kadams54 is now known as kadams54-away === kadams54-away is now known as kadams54 === kadams54 is now known as kadams54-away [20:43] yuilibrary.com is down, I pinged some of the guys on the old YUI team to look into it - we may not have YUI docs for a bit....who knows [20:43] heh [20:51] * teslanick awaits yuidoc.juju.ubuntu.com [20:52] lol [20:53] lazyPower: have you had a chance to give the latest Ghost release a try? [20:53] hatch__: i haven't [20:53] been a bit preoccupied being out of state [20:54] ahh === kadams54-away is now known as kadams54 [21:30] somehow the ac removal has caused the inspector tests to break :/ [21:31] hatch: lol [21:31] I do not get it, it's very confusing [21:33] hatch: let me know if you need another set of eyes === perrito6` is now known as perrito666 [21:36] ok well I figured out what the problem is, but I am not sure why it started now. [21:37] query selectors which are querying on things like [name=mediawiki/7] or [data-bind=config.admin] === kadams54 is now known as kadams54-away [21:37] change it to be [name="mediawiki/7"] or [data-bind="config.admin"] [21:38] and they pass [21:40] hmm, no idea [21:40] yeah just switched back to develop and they all pass [21:40] I'm going to checkpoing this commit and push it up [21:40] maybe some fresh eyes can see what's up [21:42] rick_h_: https://github.com/juju/juju-gui/pull/669/files [21:42] if you get a moment === kadams54-away is now known as kadams54 [21:48] hatch: race? is databinding all in place? [21:48] hatch: but yea nothing jumps out at me [21:49] well the issue is that it can't find the element so I debugger it and indeed it can't [21:49] I wrap the names and such in " and it works [21:49] but there are hundreds of places where this is done [21:49] oh, huh strange [21:49] AND it works fine in develop... [21:50] is container different at all? [21:50] from develop to this, exact same node/path? [21:50] yeah - it's not that the container doesn't exist - it's that the query for it returns undefined if they aren't quoted [21:51] it makes a little sense because they are special characters..but I'm pretty sure it was valid [21:51] hatch: yea, I'm wondering if there's some diff in native dom element type or something before/after the code change? [21:51] but yea, no idea off the top of my head [21:51] I think I'll start a new branch and remove code/run tests [21:52] and see what triggers it [21:52] ok [21:52] 1000+loc code removal is sure gona be nice though :) [21:52] hey! that's more of my code! :P [21:53] lol [21:54] who knows maybe there will be some revolutionary blog post come out of this [21:54] ...or I deleted a comma somewhere... [21:54] :P [22:13] Morning [22:24] hey huwshimi === kadams54 is now known as kadams54-away === kadams54-away is now known as kadams54 [22:34] ok the failures were caused by removing the 'autocomplete' module [22:34] which is not used anywhere in the codebase [22:34] ugh === kadams54 is now known as kadams54-away [22:57] uiteam lf reviews and qa on https://github.com/juju/juju-gui/pull/670 [22:57] On it [22:58] thanks [22:58] doh [22:58] lint error [22:58] sorry will fix [22:59] update [22:59] d