#clojure – 2013-02-07

2013-02-07

00:17	Deece	Sorry if this is OT for #clojure, but I read on the google group a while ago the mention of a new-ish (>2009?) lisp unrelated to Clojure. it had pattern matching. it was called something er, vaguely "buddhist"-sounding?
00:18	Deece	Wondering if it rings a bell for anyone.
00:19	technomancy	there's Qi, which is a lisp with interesting ideas about static types and a horrible license
00:19	technomancy	or Shen rather
00:19	alisdair	joxa or lfe? the erlang lisps? they have pattern matching
00:21	Deece	technomancy: it was Shen, thank you. The shen license looks.. weird, but not too bad.
00:23	technomancy	depends what you're going to use it for I guess
00:24	technomancy	not the kind of thing I'd invest a lot of time in if someone else is going to tell me what I can use it for
00:30	shaunxcode	what uses does the license prohibit?
00:35	technomancy	distributing anything based on Shen which doesn't pass the spec
00:35	technomancy	among other things
00:35	technomancy	kinda similar to the way oracle restricts openjdk forks with the compatibility kit
00:36	Guest76208	isnt that a good thing? J++?
00:37	amalloy	ask apache about harmony
00:37	tmciver	technomancy: it doesn't sound too restrictive if you're only creating apps with it: "the code you write which rides on the top of our code is not bound by our license, this is yours."
00:38	technomancy	Guest76208: in the case of Oracle it's very bad, since they don't let you even have the compatibility kit unless you're their pal
00:39	shaunxcode	I think the most interesting thing about shen is KL, I have been waiting to see if anyone else uses it as a target for other lisps (or languages in general)
00:39	technomancy	Guest76208: in the case of Shen, it's bad because they won't even let you base your work on it if you change the name. it's not like they're just out to avoid confusion or anything.
00:40	technomancy	if you read the reasoning behind the license, you can tell he's just bitter about his inability to make a living from the software he's written
00:40	technomancy	Oracle's is worse because if you get on their bad side they've got a big pile of patents they'll use against you, of course. Shen is just silly.
00:41	technomancy	(silly about compatibility)
00:42	shaunxcode	yeah the notion of anything beside MIT licensing just seems odd to me but I'm a simple kind of guy
00:42	shaunxcode	(for open source/languages that is)
00:42	technomancy	ugh; MIT's a bit of a mess when it comes to patents though
00:42	technomancy	someone can submit a patch to your MIT-licensed codebase, get bought by oracle, and then sue your pants off
00:43	technomancy	most licenses longer than a page avoid that
00:44	shaunxcode	hah really? I guess I should pay more attention - is their a case where something like that has happened?
00:45	technomancy	not sure if it's happened with MIT specifically, but it has no patent provisions at all, which is foolish for anything high-profile.
00:47	amalloy	does anyone know if every contrib requires patches come exclusively through jira? ISTR rich saying at some point that as long as you only take patches from people with CAs you're okay
00:48	amalloy	i ask because i have some small fixes to tools.reader and don't fancy sticking my hand into the jira bag of scorpions
00:57	shaunxcode	anyone have thoughts on how to pretty print a map with a literal tag in it e.g. for creating datomic schemas programatically? I have a quaint work around, but now I am intrigued as to how you can do it
01:04	callenbot	yogthos: next time you're around, I have some potential additions/improvements to Luminus to discuss.
01:04	callenbot	yogthos: I'd just do a PR, but I'd rather discuss them with you first
01:18	fsvehla	I have a deconstructuring question, is it possible to get a map with {key, (the-vector 1), other-key (the-vector 2), ... from a vector?
01:18	Ember-	the-vector meaning element at index x?
01:18	fsvehla	I currently use { :host (fields 0) :timestamp (fields 1) :url (fields 2) ...
01:19	fsvehla	Ember-: Yes
01:19	fsvehla	Of course starting with 0
01:20	fsvehla	I was looking over the destructuring examples of jay fields, but I don’t actually think that’s possible...
01:20	amalloy	(zipmap [:host :timestamp] v)
01:21	fsvehla	Actually I think locals following the let form would be even nicer
01:21	fsvehla	zipmap, thanks
01:22	Ember-	I'm not sure if that's possible via destructing
01:22	Ember-	it's a one liner otherwise though
01:24	Ember-	,(map-indexed hash-map ["a" "b" "c" "d"])
01:24	clojurebot	({0 "a"} {1 "b"} {2 "c"} {3 "d"})
01:24	Ember-	wasn't that what you wanted?
01:24	Ember-	no it was not
01:24	Ember-	almost!
01:26	fsvehla	The zipmap solution works to return a map with the specified keys :)
01:27	fsvehla	I just wasn’t sure if that isn’t possible via destructing, because it is a kind-a obvious feature
01:33	amalloy	fsvehla: nonsense. destructuring isn't for building objects, it's for tearing them apart. and you asked to build a map
02:09	durka42	what on earth does that mean
02:09	durka42	"PLASMA warning: plasma_init(): PLASMA re-initialized"
02:09	durka42	(using incanter)
04:03	AtKaaZ	has anyone found a solution for this or is it impossible? &(#(var %) prn)
04:03	AtKaaZ	&(#(var %) prn)
04:03	lazybot	java.lang.RuntimeException: Unable to resolve var: p1__12841# in this context
04:04	AtKaaZ	&(#(var-get %) prn)
04:04	lazybot	java.lang.ClassCastException: clojure.core$prn cannot be cast to clojure.lang.Var
04:04	durka42	wait what are you trying to do
04:05	Sgeo	var-get should be a function like any other
04:05	AtKaaZ	to be able to give a symbol to a function and get the var, if that even makes sense
04:05	Sgeo	resolve?
04:05	clojurebot	Unable to resolve symbol: pst-elem-str is you have an old version of clj-stacktrace somewhere. see the Troubleshooting section of the swank-clojure readme.
04:05	Sgeo	(doc resolve)
04:05	clojurebot	"([sym] [env sym]); same as (ns-resolve ns symbol) or (ns-resolve ns &env symbol)"
04:05	Sgeo	(resolve 'resolve)
04:05	Sgeo	,(resolve 'resolve)
04:05	clojurebot	#'clojure.core/resolve
04:05	AtKaaZ	something like: (fn [func] (add-hook func #'incTimes_hook))
04:06	durka42	,(class #'prn)
04:06	clojurebot	clojure.lang.Var
04:07	AtKaaZ	where func needs to be a var
04:07	AtKaaZ	but I don't want to pass a var, I want to pass a symbol
04:07	Sgeo	AtKaaZ, what's wrong with resolve?
04:07	AtKaaZ	Sgeo, don't know how to use it in this context
04:07	Sgeo	If what you're doing is even sensible, which I don't know
04:08	Sgeo	Pass it a symbol, it finds the var that would be named by the symbol in the current namespace
04:08	AtKaaZ	oh it has to be quoted, wait
04:11	AtKaaZ	&(#(resolve (quote %)) prn)
04:11	lazybot	java.lang.SecurityException: You tripped the alarm! resolve is bad!
04:11	AtKaaZ	i don't know how:)
04:13	AtKaaZ	maybe from a macro, trying...
04:14	durka42	you'll need a macro, by the time #(resolve (quote %)) gets prn it has already been resolved into #'clojure.core/prn
04:14	durka42	,'(prn)
04:14	clojurebot	(prn)
04:14	durka42	hmm, maybe not?
04:15	AtKaaZ	my problem is, I think, that it's passing p1__9968# to the macro, by I want its value instead
04:16	AtKaaZ	so instead of p1__9968# it would pass prn
04:16	AtKaaZ	i wish
04:19	AtKaaZ	in other words, I want the symbol which was passed to the anon function, but I can't unless the function is a macro, i see
04:20	AtKaaZ	ie. (xx prn) when xx is a macro will get me prn inside the macro, but (#(xx %) prn) will never get me "prn" but only things like p1__13332# or #<core$prn clojure.core$prn@741854be>
04:23	AtKaaZ	unless there's a way to get "#<core$prn clojure.core$prn@741854be>" to "prn" ?
04:23	AtKaaZ	,prn
04:23	clojurebot	#<core$prn clojure.core$prn@37f8ff58>
04:24	AtKaaZ	that's one way, but I want the opposite way
04:29	AtKaaZ	ok, I got what I wanted for, thanks , if anyone cares it's this: https://github.com/DeMLinkS/demlinks/blob/92379ca7f955033d18a3b8eec677ce1cc1e7db25/src/runtime/q.clj#L673
04:31	AtKaaZ	(ignore that :pre btw xD)
04:37	AtKaaZ	right that makes sense why you can't resolve backwards from #<core$prn clojure.core$prn@37f8ff58> to prn because more than one symbol maybe be associated with that #<core$prn clojure.core$prn@37f8ff58>
04:40	OscarZ_	if i want to define a function with multiple arities, do I have to use defn or is it possible with just fn ?
04:53	AtKaaZ	OscarZ_: you can use fn
04:53	nonuby	i dissoc a key from a map and save that into mongo via monger, mongodb has the document with the key set to null, is there an alternative to dissoc
04:56	nonuby	ignore i was doing something stupid
04:58	nonuby	can i use update-in to update multiple keys
05:40	lpetit	Hello there
05:40	cemerick	hi :-)
05:40	lpetit	What is the best available library for file manipulation / folder creation, etc., etc. over there ? (beyond plain use of java API)
05:40	lpetit	cemerick: you're everywhere :-)
05:41	cemerick	maybe Raynes' fs lib?
05:41	cemerick	I've never used it
05:41	lpetit	or maybe just plain old apache commons io ?
05:49	ejackson	lpetit: I've been using Raynes' fs
05:49	ejackson	its good
05:50	lpetit	ejackson: thanks for the feedback. And I also see that it has apache commons compress as a dependency, so hopefully it will provide me with not so low level APIs for e.g. unzipping files
06:13	pepijndevos	If the default lein template includes a .gitignore, why doesn;t it come with the git repo right away?
06:13	durka42	you might be like me and create the lein project inside a larger extant git repo
06:17	pepijndevos	hm
06:20	pepijndevos	I like the idea of drip, but it doesn't seem to help me much. lein repl still takes several seconds to start.
06:21	pepijndevos	In fact, there is not decernible difference on either my mac or linux laptop. drip ps does show a ton of vms running. I don't get it.
06:30	the-kenny	pepijndevos: I remember problems with drip & lein2
06:30	pepijndevos	ew
06:30	the-kenny	lein2 used some caches which changed the command line of the JVM so drip couldn't cache
06:30	the-kenny	dunno if that was fixed
06:31	pepijndevos	oh, it says you need to build it yourself
06:31	pepijndevos	all the way down: https://github.com/flatland/drip/wiki/Clojure
06:31	pepijndevos	oh, the other way around
06:31	pepijndevos	you need to build leiningen
06:32	clgv	pepijndevos: well latest snapshot after preview 10 should be included in 2.0.0
06:33	aroemers	pepijndevos: Also make sure you have the latest version of drip. With some earlier version, drip had a bug that it did not reuse any spawned VM.
06:34	pepijndevos	bottom line: version hell
06:39	pepijndevos	oh, I have lein 2.0.0--SNAPSHOT, and a pretty old one at that, I guess
06:39	pepijndevos	$ lein upgrade
06:39	pepijndevos	The upgrade task is only meant for stable releases.
06:51	clgv	pepijndevos: install leiningen manually
07:01	pepijndevos	hm, the brew seems to be cray broken
07:01	pepijndevos	stable is 1.7, devel is 2.0.0 and HEAD is 2.1.0-SNAPSHOT, which is not on S3.
07:09	AtKaaZ	is there some testbed that causes my function f that I pass to swap! to be call more than once? i need to test that somehow and see what happens
07:09	pepijndevos	grrr, I'm now on a checkout version of drip and the latest leiningen that I can get. Nothing
07:10	AtKaaZ	pepijndevos: is that lein from github master branch?
07:11	pepijndevos	AtKaaZ: no, just 2.0.0 :(
07:12	pepijndevos	there is no such thing as 2.1.0-SNAPSHOT on amazon
07:12	AtKaaZ	no idea how that works, I take it you can't compile it there?
07:13	AtKaaZ	~logs
07:13	clojurebot	logs is http://clojure-log.n01se.net/
07:13	pepijndevos	I suppose I can. I don;t know. It refers to a nonexsistant hacking section of the readme
07:14	clgv	pepijndevos: lein 2.0.0 is stable
07:14	AtKaaZ	maybe you wanted this: https://github.com/technomancy/leiningen/blob/master/CONTRIBUTING.md#bootstrapping
07:17	pepijndevos	clgv: so do I need master for drip?
07:18	clgv	pepijndevos: take that on: https://raw.github.com/technomancy/leiningen/stable/bin/lein
07:18	pepijndevos	clgv: that is what I have
07:19	clgv	pepijndevos: if that does not work you then maybe something for drip was broken again in 2.0.0
07:20	pepijndevos	I'm going to try lein trampoline, that always fixes stuf… oh, no, maybe not
07:21	AtKaaZ	if anyone cares, on win7 this is what I use to build leiningen from master https://gist.github.com/AtKaaZ/4730559
07:27	wunki	how would I get the metadata of a namespace. `(meta ns) keeps returning `nil`
07:27	AtKaaZ	,(meta (var ns))
07:27	clojurebot	{:ns #<Namespace clojure.core>, :name ns, :added "1.0", :doc "A clojure.lang.Namespace object representing the current namespace.", :tag clojure.lang.Namespace}
07:28	AtKaaZ	that's not it xD
07:29	wunki	AtKaaZ: nope :(
07:33	clgv	try ##(meta (the-ns 'clojure.set))
07:33	lazybot	java.lang.SecurityException: You tripped the alarm! the-ns is bad!
07:36	pepijndevos	<inappropriate terms> I'm now running checkout versioins of both drip and leiningen. Still takes like ten seconds to do lein repl, while having a dozen jvms running, each consuming over 50MB of RAM
07:39	AtKaaZ	wunki: that actually works if the ns has meta, (meta ns)
07:40	AtKaaZ	wunki: like https://www.refheap.com/paste/10994
07:40	wunki	AtKaaZ: I think I found my problem, maybe the `ns` doesn't get recompiled in Emacs. A fresh `lein run` seems to find the meta now
07:41	AtKaaZ	wunki: yes I had to restart my ccw repl also
07:44	wunki	AtKaaZ: thanks for checking it out
07:45	AtKaaZ	wunki: np, I enjoyed it
07:46	pepijndevos	wut, where did my Java prefs go in OS X 10.8?
07:47	pepijndevos	uh, found it.. It's in system prefs, but it's not
07:51	pepijndevos	wut, no more java 6 I guess
07:53	pepijndevos	or, just chaos. I can no longer select the version, like I could in the utilities java thing. the prefs say 7, the terminal says 1.6
07:57	pepijndevos	http://reviews.cnet.com/8301-13727_7-57533880-263/java-preferences-missing-after-latest-os-x-java-update/
08:35	TimMc	I wonder if core.logic could be used to solve this puzzle: http://www.coinheist.com/rubik/a_regular_crossword/grid.pdf (PDF warning)
08:39	clgv	TimMc: hmm you would have to normalize it I think
08:40	clgv	TimMc: you can use the regexps as predicates but you have to determine the domains for the fields
08:45	clgv	TimMc: but a real unification would be difficult I guess
08:49	progo	did some of the clojure books have a venn diagram of clojure's essential sections I could use in a presentation?
08:50	hyPiRion	progo: Essential sections?
08:51	hyPiRion	http://clojure.com/ has a diagram with focus, empowerment and simplicity.
08:51	abp	Oh, great: https://codeq.io/
08:52	progo	hyPiRion: like that but less enterprisey, and in a book that has better authority in citations :P
08:54	noidi	progo, the joy of clojure has one
08:54	noidi	on page 4
08:55	progo	noidi, great. I was just at page 16 walking backwards :D
08:56	noidi	:D
09:22	octagon	hello, does anyone know of a document that describes how the case* special form works? i see it in the macroexpansion for case, but it's very confusing and the java code is sort of without comments
09:24	octagon	https://github.com/clojure/clojure/blob/master/src/jvm/clojure/lang/Compiler.java#L8362 <-- that is the java implementation
09:26	octagon	https://gist.github.com/micha/4731233 <-- macro expansion of a case expr
09:34	luxbock	is there a way to auto-load some code when nrepl starts itself in Emacs?
09:35	luxbock	for example I would like to have Pomegranate loaded everytime by default in case I need to install any new libraries
09:44	@rhickey	When we support #bin"base64" - which base64 and why? http://en.wikipedia.org/wiki/Base64
09:45	`fogus	oh my... I had no idea that there were so many choices.
09:48	@rhickey	http://tools.ietf.org/html/rfc4648 ?
09:48	@rhickey	Is there a Java lib that does rfc4648?
09:49	hyPiRion	$google java rfc4648
09:49	lazybot	[RFC4648 - jinahya - RFC-4648 Implementation. - Jinahya's Java ...] http://code.google.com/p/jinahya/wiki/RFC4648
09:55	stuartsierra	Apache Commons Codec is probably the most commonly-used Java lib, it claims to comply with RFC 2045.
09:55	jkkramer	other std libs: ruby - rfc 2045, python - rfc 3548, php - rfc 2045
09:55	@rhickey	stuartsierra: yeah, but http://www.ietf.org/rfc/rfc2045.txt looks gross
09:55	jeremyheiler	what does data.codec implement?
09:56	aroemers	rhickey: whatever the choice, I am happy to see #bin will become part of edn. :)
10:00	@rhickey	I like filename/url compatible and no fixed line lengths
10:00	clgv	ssh seems to use RFC2045 for keyfiles
10:00	@rhickey	no fix/max
10:00	TimMc	rhickey: Strong preference for +/
10:01	@rhickey	TimMc: because?
10:01	TimMc	It's the original, so basically every base64 tool I've seen uses it.
10:02	TimMc	The URL/Filename stuff is just... I don't know, a special transport-dependent encoding.
10:02	@rhickey	and yet, most newer ones don't, for good reasons
10:02	@rhickey	transport-dependent? transport compatible
10:02	TimMc	Sure.
10:03	TimMc	Could you support both? It doesn't look ambiguous.
10:03	TimMc	One danger is that you would misread (say) _- variants instead of rejecting them.
10:04	TimMc	(because it uses the same alphabet as -_)
10:04	hyPiRion	Wouldn't it be smarter to use the standard used by most other languages? As long sound implementations of the standard exists, it will be both easier and simpler to extend/implement EDN on those platforms then. Unless there's significant issues with the most popular, I don't see a reason why this should be a discussion.
10:04	@rhickey	hyPiRion: "I don't see a reason why this should be a discussion." - then don't participate
10:04	TimMc	That's rude.
10:05	@rhickey	I don't see a reason why this should be a discussion. is rude
10:05	TimMc	I think he's asking you to justify the bikeshedding.
10:05	@rhickey	I think the existence of the variants demonstrates the bikeshedding is not due to me
10:06	TimMc	You expressed a preference for filename/URL compatible variants -- why?
10:06	pepijndevos	technomancy: ping
10:07	hyPiRion	rhickey: You lost the head of that sentence. What I meant was that if there's minor differences between the characters used, then use the most popular one.
10:07	hyPiRion	Disclaimer: I have no knowledge on base64.
10:15	aroemers	One could always add a #my/rfc2045 data reader oneself of course, if the default is not to one's liking.
10:16	TimMc	#base64-_
10:16	TimMc	(not entirely serious)
10:17	TimMc	,(javax.xml.bind.DatatypeConverter/printBase64Binary (.getBytes "k\ueff3"))
10:17	clojurebot	"a+6/sw=="
10:23	pepijndevos	Is (:require :refer) the same as (:use :only) ?
10:28	saolsen	pepijndevos: yep, and :refer :all would be the same as just :use
10:29	pepijndevos	I was hoping I could use slamhound to develop with :use and publish with :only/refer
10:29	pepijndevos	That way, a reader knows where everything comes from, and I don't have to spell it out
10:30	hyPiRion	(:require [foo :refer [bar]]) would make it possible to use foo/baz as well
10:35	gfredericks	what's a use case for evaling IFns? i.e., why does (eval (list + 1 2 3)) work?
10:36	nDuff	gfredericks: Sorry? (list + 1 2 3) is just a PersistentList, not an IFn.
10:36	danielglauser	hyPiRion: How so? Just tried at the repl, if I :refer [bar] I can't access anything else from foo besides bar
10:37	hyPiRion	danielglauser: what?
10:37	hyPiRion	,(require '[clojure.pprint :refer [cl-format]))
10:37	clojurebot	#<ExecutionException java.util.concurrent.ExecutionException: java.lang.RuntimeException: Unmatched delimiter: )>
10:37	hyPiRion	,(require '[clojure.pprint :refer [cl-format]])
10:37	clojurebot	nil
10:38	hyPiRion	,(cl-format true "~R" 150)
10:38	clojurebot	one hundred fifty
10:38	hyPiRion	,(clojure.pprint/pprint {:a 1 :b nil})
10:38	clojurebot	{:a 1, :b nil}\n
10:38	danielglauser	hyPiRion: Got it, didn't realize you were talking about using the fully qualified namespace
10:39	hyPiRion	oh
10:39	Sgeo	http://ideone.com/O0LqJy
10:39	Sgeo	Huh. That's simpler than I expected
10:41	gfredericks	nDuff: yes but + is an IFn, and it's not quoted
10:41	clgv	Sgeo: what?
10:41	Sgeo	clgv, using deftype to store data
10:41	Sgeo	(I need it to be a type to distinguish from other kinds of data)
10:42	clgv	Sgeo: you can also use maps with :type metadata for that ^^
10:42	Sgeo	I'm just writing a version of trampoline which requires the loop to be ended with (done blah) rather than just blah, meaning functions can be returned
10:42	Sgeo	It's for my blog post
10:42	clgv	&(type (with-meta {:a 1} {:type :blubb-type}))
10:42	lazybot	⇒ :blubb-type
10:43	Sgeo	That argues that that sort of thinking is a good thing
10:43	nDuff	gfredericks: eval would be confusing to use if it didn't behave that way. Would you _really_ want to force folks to pass '+ instead of +?
10:45	Sgeo	IFns functioness is only meaningful at the beginning of a list (or when something calls it, say, if you're passing it in to a higher order function)
10:45	hyPiRion	Well..
10:45	Sgeo	(+ 1 2 3) this, as a list, sees + at the beginning is an IFn, so evaluates it
10:45	clojurebot	#<RuntimeException java.lang.RuntimeException: java.lang.NumberFormatException: For input string: "+">
10:45	Sgeo	..?
10:45	Sgeo	(+ 1 2 3)
10:45	clojurebot	suffusion of yellow
10:45	hyPiRion	Sgeo: no, that resolves the symbol
10:45	hyPiRion	,('+ 1 2)
10:45	clojurebot	2
10:45	clgv	(+ 1 2 3),
10:45	clojurebot	suffusion of yellow
10:45	clgv	oops
10:46	clgv	,(+ 1 2 3)
10:46	clojurebot	6
10:46	hyPiRion	(eval (list 'foo 1 2)) returns "unable to resolve symbol: foo in this context"
10:46	Sgeo	Well, yeah, symbols get resolved first
10:46	hyPiRion	Whereas double-quoting it resolves that issue
10:47	Sgeo	What's the problem exactly?
10:47	hyPiRion	Sgeo: I have no issue with it, really. Behaves as I expect.
10:47	Sgeo	Someone seemed like they were having an issue?
10:47	Sgeo	Don't mind me, I'm tired
10:48	Sgeo	And forcing myself to stay awake so I can stop sleeping in the middle of the day
10:48	hyPiRion	Sgeo: yeah, gfredericks mentioned something I didn't quite grasp
10:49	Sgeo	<gfredericks> what's a use case for evaling IFns? i.e., why does (eval (list + 1 2 3)) work?
10:49	Sgeo	gfredericks, all Clojure code is data
10:50	Sgeo	First it's read in, which turns "(+ 1 2 3)" for example into a list containing a symbol +, a number 1, a number 2, a number 3
10:50	Sgeo	Then that list is evaluated
10:50	Sgeo	(Or whatever the result of reading it is)
10:50	Sgeo	,(read "(+ 1 2 3)")
10:50	clojurebot	#<ClassCastException java.lang.ClassCastException: java.lang.String cannot be cast to java.io.PushbackReader>
10:50	Sgeo	:(
10:50	Sgeo	(doc read)
10:50	clojurebot	"([] [stream] [stream eof-error? eof-value] [stream eof-error? eof-value recursive?]); Reads the next object from stream, which must be an instance of java.io.PushbackReader or some derivee. stream defaults to the current value of in ."
10:50	hyPiRion	read-string
10:50	Sgeo	,(read-string "(+ 1 2 3)")
10:50	clojurebot	(+ 1 2 3)
10:51	Sgeo	That + is a symbol, which then gets resolved, in this case, to a function
10:51	hyPiRion	Ooh. But (list + 1 2 3) doesn't do that. It resolves + into the function plus.
10:51	the-kenny	,(list + 1 2 3)
10:51	clojurebot	(#<core$_PLUS_ clojure.core$_PLUS_@463bb30a> 1 2 3)
10:51	Sgeo	Ah, so, in that case, eval is seeing the function object directly, rather than the symbol
10:52	the-kenny	,(#<core$_PLUS_ clojure.core$_PLUS_@463bb30a> 2 3 4)
10:52	clojurebot	#<ExecutionException java.util.concurrent.ExecutionException: java.lang.RuntimeException: Unreadable form>
10:52	Sgeo	Well, that's part of what evaluation means
10:52	Sgeo	Evaluation on a symbol resolves the symbol and evaluates the result
10:52	Sgeo	Erm, wait, it... hmm
10:53	Sgeo	So resolving is not theoretically a separate step from evaluation, it's what it means to evaluate a symbol
10:53	Sgeo	although, hmm. (list 'quote 'foo) gets evaluated to foo but the foo symbol doesn't get evaluated
10:54	Sgeo	I guess quote says "don't evaluate this" and symbols say "go ahead and evaluate the result"
10:54	hyPiRion	Sgeo: I think it's because functions are values.
10:55	Sgeo	It bothers me though that Clojure doesn't accept arbitrary objects in forms
10:55	hyPiRion	(= (eval 1) (eval '(quote 1))), same applies for functions
10:56	hyPiRion	A value resolves to itself, whereas a symbol doesn't.
10:56	Sgeo	(def foo 'foo)?
10:57	ivaraasen	hyPiRion: ever tried test.generative BTW?
10:57	Sgeo	I think there's some sort of phases mechanism that I'm forgetting about
10:58	Sgeo	Or... not quite phases
10:58	hyPiRion	Sgeo: You meanie. A symbol resolves to the value it refers to in the context it is evaluated in.
10:58	hyPiRion	ivaraasen: Not generative, I'm afraid
10:58	hyPiRion	just tesrt
10:58	hyPiRion	*test
10:59	ivaraasen	hyPiRion: pretty sweet stuff. using it to test array-utils.
10:59	hyPiRion	ivaraasen: What is its use case?
10:59	hyPiRion	Does it generate data to work with?
10:59	hyPiRion	/s/work/test/
11:00	clgv	hyPiRion: there are two steps: resolve and var-get (or calling the symbol)
11:00	clgv	I meant: *the var
11:00	ivaraasen	hyPiRion: yeah, in my case it just generates tons of different long-arrays and double-arrays to test the rigidity of the exposed API
11:01	Sgeo	...I forgot about vars
11:01	ivaraasen	so things that shouldn't be mutated aren't mutated, et cetera.
11:01	Sgeo	derpderpderpderpderpderp
11:01	hyPiRion	ivaraasen: Oh, that sounds valuable.
11:01	hyPiRion	At least for some stuff.
11:02	ivaraasen	hyPiRion: discovered a bug in afill! (and afill-bounded!) thanks to this, actually.
11:14	gtrak	anyone have a working phonegap/android clojurescript repl or know of a technical reason why it wouldn't work? (eg, does phonegap or android disallow eval or something weird like that?)
11:16	matthavener	gtrak: its my understanding that android doesn't preclude clojure, but the gc pressure is a little high for the dalvik jvm
11:16	gtrak	ah, yes, that's why I'm considering clojurescript as an alternative. Real clojure would be nice, but impractical.
11:17	cemerick	gtrak: clojurescript does not have eval, unless you're talking about kanaka's fork
11:17	gtrak	I see a bunch of getting-started things, but no browser-repls on google
11:18	gtrak	cemerick: 'eval' in the sense of browser-repl, doesn't it rely on JS eval?
11:18	cemerick	no
11:18	gtrak	ah
11:18	cemerick	The ClojureScript compiler is implemented in Clojure.
11:18	gtrak	I get that, I'm talking about the browser repl
11:18	gtrak	it somehow sends forms over the wire and runs them
11:19	cemerick	The browser-repl is just a different runtime environment for cljs; it still needs a cljs compiler.
11:19	pimeys	btw. let's build a browser with llvm support :)
11:19	gtrak	yes, I have the cljs compiler part working :-)
11:19	pimeys	just came to my mind while I'm writing javascript here again :(
11:19	gtrak	not the browser-repl part (works locally, not on phonegap)
11:20	alexnixon	how is Clojure (the language itself) built and deployed? Does it come from a CI machine or is it built and pushed manually on someone (rhickey's?) personal machine?
11:20	cemerick	gtrak: You might want to check out kanaka's fork, he's been working on it for quite a while now (and perhaps using it in real stuff?)
11:20	gtrak	ie, I've compiled a cljs file into javascript that should connect to lein trampoline cljsbuild repl-listen, but silently breaks
11:21	cemerick	alexnixon: see http://build.clojure.org
11:21	cemerick	gtrak: Whatever the issue there, you'll still need a Clojure runtime to compile cljs to drive a browser-repl.
11:22	gtrak	that's what lein-cljs repl-listen does
11:25	alexnixon	cemerick: thanks for that - just trying to figure out the best way of managing build/releases internally
11:26	cemerick	gtrak: oooh, you just want to run browser-REPL with cljs running on android, not have a self-hosted cljs
11:26	gtrak	yes! :-)
11:26	cemerick	The joys of short-form textual communication.
11:27	cemerick	gtrak: silent failure might sound like you're violating the same-origin policy
11:27	gtrak	ah, I was just thinking it could be a security policy
11:28	gtrak	the android emulator has a hook for 10.0.2.2 to refer to the host IP, that's what address I'm trying to connect to
11:28	gtrak	when every other cljs example I've seen is localhost, (won't work in this case)
11:29	cemerick	gtrak: If you can get the same file to work when loaded from localhost (not file:/), then yeah, you're running into a phonegap/network/browser security restriction.
11:29	cemerick	You can't just throw a random URL into your browser repl connection call
11:29	gtrak	that makes sense
11:29	cemerick	it has to be the same host as where the cljs was loaded from
11:30	gtrak	I will hack it :-)
11:37	gtrak	cemerick: looks like the browser-repl indeed uses js eval? just curious now: (js* "eval(~{block})"))
11:39	gtrak	along with some crazy string interpolation macro trick
11:40	gfredericks	Sgeo: hyPiRion: I just noticed the long conversation I spawned; was there any resolution?
11:40	cemerick	gtrak: oh, cljs does use js eval, but not in service of compiling cljs->js
11:41	cemerick	which is what I thought you meant
11:42	gtrak	ah, yea, no. cljs-eval js-eval browser-repl-eval clojure closure google-closure, I think we need some more ambiguity :-)
11:42	Sgeo	My own personal conclusion is that I need more sleep.
11:46	pepijndevos	Any carmine users around? I'm confused about how it does pipelining. Am i correct that any command ran inside a car/with-conn actually returns nil and only returns the result at the end?
11:47	alexnixon	cemerick: with the Clojure release process, what is to prevent a sneaky commit being pushed by someone in the second before rhickey clicks the "release" button in Jenkins? Is it safe because rhickey is the only guy with push-to-master privileges?
11:48	cemerick	alexnixon: AFAIK, only he and Stuart Halloway have commit. That might be outdated info, but it's a very limited set in any case.
11:50	alexnixon	cemerick: thanks for the info - that sounds plausible
11:51	Chousuke	stuff like that is probably the main reason I like git
11:51	Chousuke	everyone has commit access, so you only need to decide whose collection of commits you like best
11:52	alexnixon	git definitely makes a bunch of different workflows possible
11:52	alexnixon	it's just tricky finding the "right" one for the situation
11:52	Chousuke	I find working with svn so awkward because I can't save my code before I'm ready to publish it :/
11:53	Chousuke	fortunately git-svn exists
11:54	Chousuke	though I've found that often I end up with a bunch of crappy commits and then svn dcommit them all. :P
11:56	Chousuke	that's still marginally better than having a huge bunch of code uncommitted in the working copy, however
11:59	Chousuke	And I tend to use branches so often that I find it hard to imagine working without them :P
11:59	gfredericks	I can't imagine not viewing a git repo as a graph
12:00	Chousuke	that's what it is
12:00	gfredericks	yes I mean for looking at it visually
12:00	gfredericks	the default format of `git log` seems so worthless
12:00	Chousuke	I mostly use git log -p
12:01	Chousuke	to review changes
12:01	gfredericks	I use git log --graph --all --format=\"%h[%t] %d %an--%s\"
12:02	gfredericks	those backslashes are probably unhelpful
12:02	Chousuke	branches make everything so carefree in git
12:02	gfredericks	my favorite is debugging by creating a `wtf` branch and deleting half of the code at a time until the error is isolated
12:02	nDuff	they're certainly unhelpful.
12:02	Chousuke	have a cool new idea? branch. need to fix something? branch. hard merge/rebase? branch
12:03	nDuff	Use the slashes like that and the shell will string-split the contents
12:03	gfredericks	nDuff: I needed them in the .gitconfig as an alias
12:03	nDuff	so your command becomes: git log --graph --all '--format="%h[%t]' '%d' '%an--%s"'
12:03	nDuff	Ahh.
12:04	Chousuke	the way I usually do rebases is branch -> rebase -i -> ff merge to original. sometimes with several branching steps
12:05	Chousuke	so I end up for foo-rebase-1 through 10
12:05	Chousuke	the amusing part is that you don't even have to do that manually since git keeps all the information in the reflog anyway
12:06	Chousuke	but somehow it still helps
12:08	technomancy	pepijndevos: hi
12:10	pepijndevos	technomancy: I was playing with slamhound, hoping it would allow me to turn (:use) clauses to (:require :refer) ones. So far, it just crashes
12:10	technomancy	oh, that's a drag
12:10	pepijndevos	what do you mean?
12:11	technomancy	oh, just that it's unfortunate
12:11	pepijndevos	oh
12:11	nDuff	Heh; the joys of using slang on an international forum
12:11	pepijndevos	My use case is that when reading code, I like to see where stuff comes from, but when I write code, I don't want to specify every little function.
12:12	pepijndevos	technomancy: is that something slamhound could do, if I got it to work?
12:12	technomancy	pepijndevos: oh yeah, definitely
12:12	pepijndevos	cool
12:13	hammer__	if i'm looking at a seq and want to compare an element to its previous element
12:13	hammer__	how do i do that sort of comprehension?
12:13	pepijndevos	Failed to reconstruct: #<File src/asky/core.clj>
12:13	pepijndevos	java.lang.RuntimeException: Unable to resolve symbol: == in this context, compiling:(NO_SOURCE_PATH:1:139)
12:13	technomancy	pepijndevos: ok, yeah the regex for exceptions needs tweaking
12:13	rasmusto	hammer__: partition might be of use
12:14	pepijndevos	technomancy: what does that mean?
12:14	technomancy	shouldn't be too tricky; I think it assumes missing vars are going to be compliant with the symbol name restrictions on clojure.org
12:14	hammer__	oh that's exactly what i was look for, thanks rasmusto
12:14	hammer__	(inc rasmusto)
12:14	lazybot	⇒ 1
12:14	technomancy	pepijndevos: this explains how slamhound works: http://technomancy.us/148
12:15	pepijndevos	ok, will read
12:15	technomancy	basically it looks at the message for each exception to see what var is missing
12:15	rasmusto	hammer__: (partition 2 1 someseq) is something I use a lot
12:15	technomancy	so it can search for it and add it into the candidate ns
12:15	technomancy	so it's just a matter of tweaking the regex
12:15	pepijndevos	ah, nice. So all the 2 namespaces I tried must have defeaten the regex in some way
12:16	pepijndevos	for the other one it says: nth not supported on this type: Symbol
12:16	technomancy	that's the thing about regexes... they never cover all the possibilities
12:16	craigbro	hmm
12:17	craigbro	core.logic issue
12:17	craigbro	if I do (run N ...
12:17	craigbro	will it repeat results into I get N?
12:17	pepijndevos	craigbro: no
12:17	hiredman	I don't think so
12:18	maio	OT: native english speakers - is it possible to say word raweri? :)
12:18	craigbro	hmm
12:18	hiredman	but depending on your program you can have duplicate results
12:18	pepijndevos	&(#(println %&) 1 2 3)
12:18	lazybot	⇒ (1 2 3) nil
12:18	nDuff	maio: I don't know -- it's hard to know how it should be pronounced from only seeing it spelled. :)
12:18	pepijndevos	My new favourite syntax
12:21	desertmonad	Chousuke: I'm still a bit of a git novice. At work SVN is mandated, but I am able to use git-svn. Do you know of a work flow that would let me git clone a repository that was created by git-svn?
12:22	desertmonad	I can do the clone, of course, but git svn dcommit rebases to change the log messages (afaik)
12:22	Chousuke	hmm, yeah
12:22	desertmonad	and so the git clone diverges
12:22	gfredericks	maio: depends on what you mean by "possible" I guess; looks like pronounced "rah WHERE ee"
12:23	gfredericks	but definitely looks foreign
12:23	Chousuke	desertmonad: probably easiest to have two separate git-svn clones and just work with them treating them as two SVN committers (with occasional sending of patches back and forth)
12:23	nDuff	(though it does so only because bzr changed their backend to accomodate better SVN interop)
12:23	nDuff	s/only//
12:25	desertmonad	Yeah, I've thought about doing that. My goal is to not publish to SVN if I'm in the middle of tweaking something… best I've come up with is using rsync, which is less than ideal.
12:26	gfredericks	has using println with a lazy seq always given reasonable results?
12:26	gfredericks	(as opposed to str)
12:26	hiredman	yes
12:28	ln2	Hi all. =)
12:28	ln2	Do you guys have any recommended reading other than "The Joy of Clojure"?
12:29	gfredericks	"because I hate the joy of clojure" or "because I've already read the joy of clojure"?
12:29	hiredman	read clojure.org
12:29	hiredman	http://clojure.org/documentation http://clojure.org/rationale http://clojure.org/features
12:30	hiredman	etc
12:30	ln2	"Because I just started reading The Joy of Clojure and I am excited to line something up in front of it. =)
12:30	ln2	hiredman: Thank You!
12:32	brainproxy	ln2: Clojure Programming pub'd by O'Reilly was a great read, imo
12:32	ln2	I do have one dilemma though. I've never programmed in any language before. So I guess my question should be more along the lines of. "Is there a Clojure for Dummies?".
12:32	hyPiRion	ln2: Depends. If you want to learn Clojure, then Clojure Programming is great.
12:32	brainproxy	one of the best O'Reilly books I've ever read, in terms of quality of the writings and examples, and the depth of learning
12:33	ln2	brainproxy: The one with the big bird on the front? Haha
12:33	gfredericks	has anybody heard any anecdotes of new programmers learning clojure?
12:33	ln2	Is that the same thing as horror stories?
12:34	gfredericks	:)
12:34	brainproxy	ln2: this one http://www.clojurebook.com/
12:34	Sgeo	Is Clojure really a good idea for a starting language?
12:34	hyPiRion	ln2: I am not sure if any Clojure book is "not programmed anything yet" friendly. I do know that there are some python books out there for people completely new to programming, but I can't tell you which ones to choose.
12:35	cemerick	gfredericks: borkdude is a prof IIRC; Lee Spector also teaches total noobs Clojure. shrug
12:35	ln2	I'm starting to learn the difference between Clojure and the more popular languages as I go. But again... I've never programmed in any language. (advantage?).
12:35	dnolen	Sgeo: IMO it's about as good as any other.
12:35	Natch	ln2: programming clojure 2nd edition by halloway and bedra is great. clojure programming from oreilly seems good but it has 3-4 annotations per page and is extremey annoying to read, imho..
12:35	Sgeo	dnolen, well, there's the whole Java interop that is around that could cause issues
12:35	ln2	To help I'm trying to learn Lambda Calculus also. I'm sure it will help. Probably... haha
12:35	Sgeo	Which is not the case with most languages
12:35	dnolen	Sgeo: doing anything useful in any language is pain.
12:36	rasmusto	ln2: could be partially an advantage, might make you take less notice of unique clojure features though
12:36	Natch	*extremely
12:36	ln2	Natch: Thank you! I will get that one also. =)
12:36	cemerick	Natch: sorry about that; it's O'Reilly house style, something we fought tooth and nail about, but lost
12:36	gfredericks	"If you haven't programmed before I recommend starting with clojure and the lambda calculus."
12:37	pbostrom_	ln2: you also might try some exercises on http://www.4clojure.com/ to reinforce your book learning
12:37	ln2	gfredericks: Is this sarcasm? Haha
12:37	ln2	pbostrom_: I've never seen this site! Thank you!!!
12:37	rasmusto	ln2: clojure koans are good too (on github), it's slightly more structured than 4clojure
12:37	Natch	cemerick: that's a shame
12:38	Natch	cemerick: because it seemed really nice otherwise
12:38	ln2	rasmusto: I am working through the Clojure Koans right now as we speak. =)
12:38	rasmusto	ln2: cool :)
12:39	Frozenlock	I just realized why the docstring is before the arguments in clojure, as opposed to say elisp... For when you have multiple arity.
12:39	rasmusto	Frozenlock: oo, I was wondering that too. Good point
12:40	creese	Has anyone read the new o'reilly book on cljs?
12:40	ln2	I was a little turned off aesthetically to the massive amounts of parens in Clojure until I saw lamda calculus. Maybe I should be thankful? O_o
12:40	gfredericks	ln2: a little bit. nothing wrong with being ambitious though :)
12:41	technomancy	SKI or go home
12:41	hyPiRion	ln2: If you're that far (doing Clojure Koans) and never programmed before, I'd go with Clojure Programming as a book for learning Clojure. You sound like a smart and motivated guy, so it shouldn't be a problem.
12:42	ln2	gfredericks: Maybe the next generation of programmers it will be ambitious to try and comprehend object oriented languages! We can all hope...
12:42	Frozenlock	ln2: http://www.thejach.com/imgs/lisp_parens.png
12:42	Sgeo	Fun fact: There's more to OO than you would think just by looking at the most popular languages
12:43	Sgeo	The style of you see in Java and C# and C++ is not the only one out there
12:43	gfredericks	Sgeo: just because something is rich and subtle doesn't make it good :P
12:43	pimeys	sicp teaches objects in chapter three
12:43	Sgeo	*style of oo
12:43	pimeys	and of course you can do the same with clojure
12:43	ln2	hyPiRion: It may seem that way... O.o. In reality I'm only average at logic and seriously inept at higher math (pretty good at mental). I'm not having the easiest time with visually following a fn embedded into a fn taking two fn ect. ect.
12:43	pimeys	having a function with inner functions, message passing and returning a closure
12:43	ln2	I'm really really trying to go slow. xD
12:44	Sgeo	message sending implies single-dispatch I think (but not certain)
12:44	Sgeo	I'm not especially a fan of single dispatch, but I'm not sure what I would like
12:44	pimeys	me neither
12:44	pimeys	but it's possible
12:44	pimeys	and one way of building your app
12:44	pimeys	I think that the hardest part with lisps is that it doesn't force you to anything
12:44	hyPiRion	ln2: Being smart is not the same as being knowledgable. Being smart is about learning rapidly and being open to new ideas :)
12:45	pimeys	you can use whatever style you prefer
12:45	technomancy	for most people the hardest part of learning clojure is unlearning bad imperative habits
12:45	Sgeo	pimeys, which gets annoying when you try to mix different OO systems together
12:45	Sgeo	Say in Scheme if one person is using TinyCLOS and another is using, I don't know, coops
12:45	pimeys	technomancy: but you can also use your imperative habits
12:45	pimeys	keep state etc.
12:45	Sgeo	Tcl has this problem too
12:45	pimeys	Sgeo: yep
12:46	technomancy	pimeys: arguably then you haven't really learned it =)
12:46	Sgeo	Although Snit tries to solve it by saying that anything that looks like an object is an object
12:46	ln2	For example. This Koans got me. I cheated and got the answer from a YouTube series.
12:46	Sgeo	I think
12:46	pimeys	technomancy: well, who says so :D
12:46	pimeys	lisps are not so much of functional languages, I'd say
12:46	ln2	http://tny.cz/7feb3618
12:46	dnolen	ln2: go slow
12:46	Sgeo	pimeys, depends on the Lisp
12:46	pimeys	but they have a nice way of giving you that possibility
12:46	Sgeo	Trying to categorize them all into one mold seems like a fool's errend
12:46	technomancy	actually that's not true, I judge all the time
12:47	ln2	Even worse. The explanation on YouTube was basically. "That was convoluted... lets move on now!"
12:47	Sgeo	Lisp is a style of syntax and metaprogramming, imo
12:47	gfredericks	technomancy: and I'm going to judge you either way so you can't really avoid it
12:47	ln2	O.o
12:48	ln2	I'm assuming all of you can solve that one pretty easily?
12:48	pbostrom_	creese: I read it, it's pretty light on content, but that's reflected in the price, probably worth if if you're brand new to cljs, but if you're already up and running, you won't learn anything new
12:48	Sgeo	Took me a bit more thinking than it should have, to be honest. But yeah
12:49	Sgeo	Think about what the fn is supposed to be doing. What is its role, what is its job there?
12:49	ln2	To remove words that don't meet certain conditions?
12:50	Sgeo	reduce isn't a mover, it's a thing that takes a list and gives a result. And the fn you pass into it is supposed to take a result so far and an item, and return a new item-so-far
12:50	Sgeo	*reduce isn't a remover
12:51	Sgeo	(also, saying "list" was a bad choice of words on my part)
12:51	ln2	Oh right. Because nothing can / should be "removed". Just returned.
12:51	ln2	Because "removing" is a mutation.
12:51	Sgeo	Well, there is a function that returns a copy with some stuff different in the copy. Removed if it doesn't meet a condition. But that's not what reduce is for.
12:52	dnolen	Sgeo: message sending isn't necessarily single dispatch, look at Cecil, also the whole idea behind CLOS was being able to support different OO systems under the same roof.
12:53	ln2	I suppose it's starting to make more sense now. I wish there was an easier way to format these but there really isn't because it's a single expression.
12:53	ln2	Maybe this IS what draws people to OO?
12:53	Sgeo	I think what draws people to OO is the fact that it's one of the first things they learn
12:54	ln2	Then maybe.. really.. we only have public schools to blame...
12:54	Sgeo	As far as formatting goes, that's not really much of an OO vs. functional thing. There are ways to split things up in functional languages
12:54	Sgeo	But imperative languages do admittedly tend to partition into separate statements more than having an expression
12:54	ln2	Sgeo: Really? I'm interested in that!
12:55	Sgeo	,(doc let)
12:55	clojurebot	"([bindings & body]); binding => binding-form init-expr Evaluates the exprs in a lexical context in which the symbols in the binding-forms are bound to their respective init-exprs or parts therein."
12:55	ln2	I wonder if I could reason how to split up that koans... might be a fun exercise.
12:55	Sgeo	,(let [a 5 b 6] (+ a b))
12:55	clojurebot	11
12:56	Sgeo	Other functional languages have other approaches. And you can usually drag an fn out and give it a name
12:56	Sgeo	(although not always)
12:56	enquora	anyone from prismatic here? is the Graph library Clojurescript-friendly?
12:56	Sgeo	But at least with this koan, it's just a matter of fill in the blank
12:56	Sgeo	I promise that for this one, the blanks are very simple, they won't complicate the expression at all
12:56	ln2	Binding... I was watching a talk about binding and how it do it without Macros.
12:57	Sgeo	,((fn [a b] (+ a b)) 5 6)
12:57	clojurebot	11
12:57	ln2	I don't actually know what let / binding do yet.
12:57	Sgeo	^^roughly equivalent to the above let
12:57	ln2	Wait a minute. Theres a REPL in this channel?
12:57	Sgeo	Two, although imperfect
12:58	ln2	Ha! That is amazing!
12:58	brianwong	that is amazing
12:59	brianwong	i didnt know that
12:59	brianwong	(doc let)
12:59	clojurebot	"([bindings & body]); binding => binding-form init-expr Evaluates the exprs in a lexical context in which the symbols in the binding-forms are bound to their respective init-exprs or parts therein."
12:59	brianwong	err
12:59	ln2	,(* 2 2)
12:59	clojurebot	4
12:59	brianwong	,(doc let)
12:59	clojurebot	"([bindings & body]); binding => binding-form init-expr Evaluates the exprs in a lexical context in which the symbols in the binding-forms are bound to their respective init-exprs or parts therein."
12:59	ln2	It's fast too. giggles like a schoolgirl
12:59	ln2	And it has the doc namespace!
13:00	ln2	Does an XMPP REPL exist?
13:00	technomancy	clojurebot used to have an XMPP interface
13:00	technomancy	I think it was removed though
13:00	ln2	=(
13:02	callenbot	enquora: I sincerely doubt it, and why would you want it to be?
13:03	enquora	need to build a PDF report generator to run in both browser and on server
13:04	enquora	it's a non-trivial task, just thinking now how to structure it
13:04	hyPiRion	enquora: What do you mean by "in browser"?
13:05	enquora	HTML browser
13:05	hyPiRion	Should the PDF be build by ClojureScript, or could it be sent from the server to the browser?
13:05	hyPiRion	built*
13:05	enquora	must run offline
13:06	enquora	it's a distributed system
13:07	ln2	Do most of you guys use LightTable?
13:07	enquora	the web browser app is a cached HTML app, or browser extension
13:07	craigbro	not I
13:07	craigbro	but I'm an emacs curmudgeon, so may not be representative, and is no judge of LT
13:08	TimMc	ln2: Most folks here use emacs or vim.
13:08	TimMc	LT is very young.
13:08	nDuff	nod. It's a very neat toy, and has promise.
13:09	shaunxcode	I seem to recall there was a pure js pdf gen tool floating around recently
13:09	ln2	Having zero programming background I'm not sure I could bring myself to use tools like emacs and vim. I'm really impressed by some of the features I've seen from both of them though, features I did not expect in cli apps.
13:09	TimMc	ln2: Learn one thing at a time.
13:09	TimMc	Both emacs and vim have long learning curves.
13:09	ln2	TimMc: That's the method I'm trying to use. LightTable is helping me get going fast. =)
13:10	TimMc	Good.
13:10	nDuff	ln2: If you want to be blown away, watch the Emacs Live intro video. http://vimeo.com/22798433
13:10	shaunxcode	ln2: the nice thing about using a generic text editor when starting is you learn the difference between what is language magic v.s. what is ide magic v.s. build tool wizardry
13:10	TimMc	I had to try learning emacs 3 times before it stuck; the first two times I was trying to learn a new programming language at the same time, so it sucked. :-P
13:11	ln2	One thing I would really like to have is some way of auto-correcting my parens. Light table color codes them. But I think it was emacs that actually does parens for you?
13:11	dnolen	ln2: yes paredit
13:11	rasmusto	ln2: paredit.el or paredit.vim
13:12	TimMc	~paredit
13:12	clojurebot	paredit is not for everyone, but what you need to understand ís how to become the kind of person that paredit ís for.
13:12	TimMc	:-P
13:12	shaunxcode	codemirror has paredit mode http://htmlpreview.github.com/?https://github.com/achengs/subpar/blob/master/demo/normal.html
13:12	technomancy	clojurebot: botsnack
13:12	clojurebot	Thanks! Can I have chocolate next time
13:12	ln2	Did the bot just talk? O_O
13:12	TimMc	I triggered a factoid.
13:12	technomancy	as long as it obeys the Three Laws clojurebot can do anything
13:12	ln2	Aahhhhahha!
13:12	ln2	Is the bot as opinionated as Hickey? xD
13:13	TimMc	The bot is opinionated as the people who care to feed it things.
13:13	TimMc	Also it randomly picks up quotes from the channel, which can be... odd.
13:14	ln2	It's also a Furby? O _ O
13:14	Sgeo	Paredit is awesome
13:14	Sgeo	I wish I had paredit in here
13:14	technomancy	Sgeo: (add-hook 'erc-mode-hook 'paredit-mode) ; =D
13:14	jballanc	clojurebot: litany against cons
13:14	clojurebot	litany against cons is "I must not cons. Cons is the perf-killer. Cons is the little death that brings total obliteration. I will face my cons and permit it to pass over me and through me, and when it has gone past I will turn my GC to see its path. And where it has gone there will be nothing; only I will remain."
13:14	Sgeo	technomancy, heh
13:14	craigbro	conj
13:14	hyPiRion	It replies to certain things. right clojurebot???
13:14	lazybot	hyPiRion: Yes, 100% for sure.
13:14	hyPiRion	oh, that's the wrong one.
13:15	llasram	technomancy: I just tried it, and paredit is unhappy lingering emoticons :-(
13:15	Sgeo	Close all emoticons! (:)
13:15	Sgeo	(:-)
13:15	technomancy	llasram: yeah, show-paren-mode is good enough for me
13:16	Bronsa	show-paren-mode is awesome
13:20	hyPiRion	I prefer highlight-parentheses.
13:23	aaelony	does anyone know of a tutorial or link that describes how to profile a clojure program running at the repl via yourkit ?
13:24	aaelony	or running a jar produced by lein uberjar ?
13:24	hyPiRion	aaelony: running a jar? java -jar name-of-uberjar
13:25	aaelony	hyPiRion: how to profile that running jar though?
13:25	ln2	http://tny.cz/f15ac6b4
13:25	ln2	I'm trying to understand how this "inline function" can be expressed in a way that is easier to read. What am I missing?
13:26	hyPiRion	aaelony: I'm afraid I don't know that. I saw the questions as two decoupled ones.
13:27	aaelony	hyPiRion: thanks anyways. yeah I have a jar I am running that I'd like to profile. I have an evaluation license for yourkit and trying to get it to analyze my jar.. Youkit's docs say I should do this: java -jar yjp-controller-api-redist.jar localhost 10001 capture-memory-snapshot but unsure what host and port and how it knows to analyze the uberjar running...
13:28	aaelony	http://www.yourkit.com/docs/12/help/command_line_tool.jsp
13:31	owengalenjones	a few minutes late, what's wrong with cons?
13:31	technomancy	this is pretty sweet: http://schneems.com/post/42508340989/open-source-in-your-inbox-code-triage
13:32	technomancy	owengalenjones: nothing apart from the fact that allocating memory is often at odds with performance
13:33	technomancy	nobody should be following the litany against cons unless they're neck-deep in profiling =D
13:35	owengalenjones	ok, heh hadnt seen that before and was curious
13:35	technomancy	it's mostly an excuse to quote Dune I guess?
13:36	desertmonad	the litany against fear got me through some final exams in years past :)
13:38	Sgeo	Huh. So that's where "politics is the mind-killer" came from, I guess
13:38	Sgeo	Dune reference
13:41	desertmonad	Dune is one of my all time favorites. That and Neuromancer.
13:41	technomancy	the bot's gotta have some neuromancer quotations in there somewhere, but I can't think of any off the top of my head
13:42	gtrak	isn't cons pretty obvious? cache-thrashing... java programmers know arraylist is generally always faster than linkedlist
13:44	nishant	Hi! Is there a way to comp a function with itself n times? Something with this behavior - (f(inc, 5), 0) => 5 ?
13:44	gtrak	iterate?
13:44	technomancy	,((apply comp (repeat 5 inc)) 0)
13:44	nishant	Iterate almost works, except it returns a seq of all the intermediate results too
13:44	clojurebot	5
13:44	gtrak	,(doc iterate)
13:44	clojurebot	"([f x]); Returns a lazy sequence of x, (f x), (f (f x)) etc. f must be free of side-effects"
13:45	owengalenjones	IANA java progrrammer
13:45	AimHere	(nth #(iterate inc 0) 5)
13:45	AimHere	,(nth #(iterate inc 0) 5)
13:45	clojurebot	#<UnsupportedOperationException java.lang.UnsupportedOperationException: nth not supported on this type: sandbox$eval77$fn__78>
13:45	gtrak	nishant: it's lazy, don't keep the head
13:45	nishant	Ah, thanks!
13:45	AimHere	Oops
13:45	gtrak	,(nth (iterate inc 0) 5)
13:45	clojurebot	5
13:46	gtrak	not sure why returning a seq would be anything other than convenient
13:46	technomancy	clojurebot: oinc
13:46	clojurebot	oinc is octo-inc: (apply comp (take 8 (repeat inc)))
13:46	technomancy	hm; that must predate repeat's second arity
13:48	dnolen	`fogus: ping
13:49	brainproxy	I must not fear. Fear is the mind-killer. Fear is the little-death that brings total obliteration. I will face my fear. I will permit it to pass over me and through me.
13:57	ln2	I am trying to re-write one of the Clojure Koans from section 8. What is going on here? Haha
13:57	ln2	http://tny.cz/6ad1b8bc
13:58	gtrak	ln2: you need to return an accumulator
13:58	gtrak	(reduce (fn [acc next-element] do-something-to-acc) source-seq)
13:59	ln2	"source-seq" is the accumulator?
13:59	gtrak	acc
13:59	ln2	Sorry for all of the questions! You guys have been a massive help. =D
13:59	gtrak	that's just kinda what reduce does :-)
14:00	ln2	The Koans are amazing but I wish there was a detailed explanation of each problem somewhere. Not a cheat sheet but just some information. xD
14:01	gtrak	well, the doc for reduce is helpful ##(doc reduce)
14:01	lazybot	⇒ "([f coll] [f val coll]); f should be a function of 2 arguments. If val is not supplied, returns the result of applying f to the first 2 items in coll, then applying f to that result and the 3rd item, etc. If coll contains no items, f must accept no arguments a... https://www.refheap.com/paste/11004
14:05	ln2	If I wanted to make this two functions then I could create one that takes an input and does a reduce with an acc. Then call that function from another function that checks if one word is larger than another?
14:10	ln2	crickets
14:11	gtrak	it is two functions, whether they're both bound to a vars or not
14:12	gtrak	but I think what you described seems inverted
14:12	TimMc	&(reduce (fn [accum next-thing] (str accum ", " next-thing)) "0" [1 2 3 4])
14:12	gtrak	reduce is the control-flow, it's calling the aggregator or whatever you wanna call it
14:12	lazybot	⇒ "0, 1, 2, 3, 4"
14:17	ln2	Thanks a lot guys! I think I'm managing to figure it out through trial and error. I'm not sure this would be possible without a REPL. Haha
14:17	jimduey	brainproxy: LOL
14:17	ln2	I have it reduced to a "key must be an integer" error. Hmm.
14:17	ln2	To google!
14:18	amalloy	ln2: you're attempting to call a vector as a function, and passing it a non-integer
14:19	ln2	amalloy: I think I'm trying to pass a vector TO a function. I wonder how I managed to pass it AS a function. O.o
14:21	TimMc	,([:a :b :c] 0)
14:21	clojurebot	:a
14:27	ln2	Alright I think my problem is that I have two functions "lengthycheck" which uses the if statement to determine which is longer (a or b). "lengthyness" uses the reduce function. The values a and b are being passed to "lengthyness" but not to "lengthycheck". I need to read more about embedding functions. =(
14:37	dnolen	lynaghk: ping
14:49	ln2	TimMc: I had my first extremely minor programming victory. I turned your expression into a function!
14:49	ln2	(defn first-in-list [a]
14:49	ln2	('[a] 0))
14:49	amalloy	errrrr, that always returns the symbol 'a, ln2, ignoring its input
14:50	ln2	CRAP!
14:50	TimMc	What you probably want is `first`.
14:51	amalloy	TimMc: really? it looks to me like he was writing identity
14:51	ln2	TimMc: I can't just express the function in the same simple way? ([:a :b :c] 0) ?
14:51	TimMc	amalloy: Not with the name "first-in-list".
14:51	amalloy	oh
14:52	TimMc	ln2: (fn [a] (a 0)), which works if a is a vector
14:52	amalloy	i never read function names. easier to just run them on all possible inputs in my head, observe the outputs, and derive the function's meaning thereby
14:53	ln2	TimMc: Holy sh* that did it! My function was calling [a] the vector and not a the value. Thanks!
14:53	svedubois	How I can write these 2 lines:
14:53	svedubois	(show image1)
14:53	svedubois	(show image2)
14:53	svedubois	in only 1 line?
14:53	svedubois	(show image1 image2)
14:54	ln2	This has the behavior I was looking for.
14:54	ln2	(defn first-in-list [a]
14:54	ln2	(a 0))
14:54	ln2	Thanks guys! I'm not as proud now though because Tim solved it for me. =(
14:54	rasmusto	ln2: ,(first [1 2 3])
14:55	ln2	rasmusto: I figured that there was an easier way to do this. I am just trying to understand the language so I'm rebuilding various expressions as functions. Thank you though!
14:55	rasmusto	,(first [1 2 3])
14:55	clojurebot	1
14:55	Guest14889	i saw recently there is a server specially for clojure, forgot the name
14:56	Guest14889	can anyone help?
14:56	ln2	In a lot of ways it is like lamda calc. Even the most basic operations has to be reduced to logic. Instead of calling + or * you need an expression.
14:56	joegallo	Guest14889: immutatnt
14:56	rasmusto	ln2: first/rest are interesting because they work on more than just one type of sequence. They'll work on vectors, maps, seqs, etc.
14:56	Guest14889	what wevserver do people commonly deploy to?
14:56	joegallo	typo
14:56	joegallo	immutant
14:56	Guest14889	joegallo: thats the one, thanks
14:56	technomancy	jetty is the most common
14:57	technomancy	but immutant can be good if you need to integrate into certain environments
14:57	ln2	rasmusto: I know about rest from the Koans problems. I'm going to write a basic rest function manually right now. =)
14:57	Guest14889	im open to anything, i want to deploy it on a new vps
14:58	amalloy	ln2: you can't really write first or rest without in fact using first and rest. they're the primitives of sequence manipulation
14:58	Guest14889	i'd probably go with jetty unless immutant offered significant benefits
14:58	technomancy	jetty is the simplest
14:59	amalloy	you can write versions that only work on vectors, as you did for first, but it's not very useful even as an exercise
14:59	akhudek	immutant is a lot more than just a web server. You may or may not need the extra features it has. Jetty is good for a straight forward web server.
15:00	ln2	amalloy: This function does not always return the first character?
15:00	ln2	(defn first-in-list [a]
15:00	ln2	(a 0))
15:00	amalloy	try it with (list 1 2 3)
15:00	Guest14889	also, is pallet the favourite for automated deployments? or do people generally use a non-clojure tool?
15:00	ln2	(first-in-list [:a :b :c]) returns :a for me
15:01	Guest14889	if anyone has any simple pallet scripts that deploy a basic app on jetty/immutant i'd be interested i seeing one
15:01	technomancy	Guest14889: pallet is a bit less polished than other tools, but the underlying models it uses are much, much better-designed.
15:01	rasmusto	,(get '(1 2 3) 0)
15:01	clojurebot	nil
15:02	Guest14889	technomancy: interesting, i'll probably give it a try at least
15:02	Guest14889	maybe i can help polish it
15:02	amalloy	ln2: so, you're saying you did not do what i suggested and try (first-in-list (list 1 2 3))
15:03	the-kenny	(list 1 2 3) is doesn't implement the same interfaces as [1 2 3]
15:04	ln2	amalloy: That fails. But "(first-in-list [1 2 3])" succeeds.
15:04	amalloy	as i said, you invented a solution that works only for vectors
15:04	technomancy	Guest14889: I'm sure they'd appreciate feedback and help. they were very responsive when I had questions.
15:05	ln2	amalloy: Interesting.
15:05	llasram	&((fn [[x]] x) (list 1 2 3))
15:05	lazybot	⇒ 1
15:05	llasram	Hehe
15:05	ln2	amalloy: However. Rest is possible with only vectors also correct? I need to use increment?
15:05	nDuff	Hugo _is_ responsive. On the other hand, the amount of ground Pallet needed to make up to cover a lot of my use cases when last evaluated was pretty overwhelming.
15:06	nDuff	Then again, if I were amenable to a nuke-and-pave cloud deployment model, that wouldn't so much be the case.
15:06	nDuff	(and it's been a long time since I took a close look)
15:07	hyPiRion	ln2: rest is possible with anything you can do `seq` on and return a sequence
15:07	hugod	nDuff: if you want incremental edits of config files, that's a hard problem
15:08	hyPiRion	,(seq (list 1 2 3 4))
15:08	clojurebot	(1 2 3 4)
15:08	hyPiRion	,(seq (vector 1 2 3 4))
15:08	clojurebot	(1 2 3 4)
15:08	hyPiRion	,(seq (hash-map 1 2 3 4)) ;; Don't worry about this for now, just be aware of it
15:08	clojurebot	([1 2] [3 4])
15:08	hugod	nDuff: or was it something else you needed?
15:09	ln2	Rest is more difficult because apparently this is impossible.
15:09	ln2	(defn rest-of-list [a]
15:09	ln2	(a (> 0))
15:10	nDuff	hugod: Not so much editing _files_ incrementally as configurations which could span multiple files / be determined only by invocation of queries or commands (example: Determining an edit list of invocations of vendor-provided tools to move from current state to desired configuration)
15:10	ln2	hyPiRion: Just noticed your response. Yes I think I do need a sequence. =)
15:10	technomancy	ln2: interesting idea. for that to work, a would have to take a predicate on the index of the element, you could do it like so: (a (partial > 0))
15:11	the-kenny	Reminds me bit of logic programming
15:11	nDuff	hugod: ...it's not an easy problem, granted, which is why I spent so much of my time a few jobs ago working on toolage to tackle it.
15:11	hugod	nDuff: so updating based on querying the current system status?
15:12	nDuff	nod.
15:12	hugod	ok, 0.8 should be much better at that
15:13	hugod	you can use the return values of actions in flow control, etc
15:13	nDuff	Nice. I'll definitely need to reevaluate, then.
15:13	hugod	should be released before clojure/west, I hope
15:14	ln2	technomancy: I'm still getting a "key must be an integer" exception. =(
15:14	technomancy	ln2: oh, I didn't mean that would work; I meant if vectors took predicates it would work.
15:15	ln2	Hahaha!
15:15	technomancy	I was noting that your assumption it did was interesting from an API design point of view.
15:15	ln2	technomancy: Well thank you very much. I try to be interesting. ;)
15:16	ln2	technomancy: I'm determined to get this to work. I'm pretty sure I need to create a sequence to index the input data and ignore the first entry?
15:17	technomancy	there are a bunch of ways you could do it; that's a decent place to start.
15:18	technomancy	assuming you have a vector, anyway
15:18	ln2	technomancy: A bunch of ways huh? Well you really know how to make someone feel smart. ; )
15:19	technomancy	ln2: well, think of it as an artistic expression more than a puzzle.
15:20	ln2	technomancy: Now I feel a ton better! If I fail I'm simply not artistic. xD
15:21	nDuff	ln2: One easy way to do that (without the obvious thing of using first, of course) would be destructuring.
15:21	technomancy	heh; I was restraining myself from saying that =)
15:22	technomancy	let him get it working this way first so he'll appreciate that more =)
15:22	lynaghk	dnolen: pong
15:22	lynaghk	dnolen: I just saw your core.logic push on github. Looking forward to checking it out---might be able to this afternoon
15:22	hyPiRion	ln2: You're looking at it wrong. You're unfamiliar with the paintbrush you use. In fact, you're unfamiliar with anything which involves paint. It's not about being artistic or not, it's about knowing the details of the paintbrush.
15:22	lynaghk	finally back in Portland
15:22	dnolen	lynaghk: some significant changes to simple unifier in master
15:23	TimMc	technomancy: And then I can show ln2 the swearjure approach to rest, and we can all appreciate the standard way more. :-P
15:23	dnolen	lynaghk: only thing left to do is to distinguish vector & sets of symbols -> constraints
15:24	lynaghk	dnolen: awesome. I'll take things around for a spin and if I'm feeling lucky I'll take a crack at the vector+sets constraints
15:24	ln2	hyPiRion: I think that is pretty much right. Compared to actual painting I'm accomplishing what amounts to a poorly drawn stick figure. Only been programming for 4 days!
15:24	ln2	nDuff: Thank you for the suggestion! I will do it that way after I figure out another way of doing it first. =)
15:24	jsabeaudry	What are the popular choices for doing parameter validation on your rest api?
15:25	hyPiRion	TimMc: yeah, I used a similar version of your rest to get quicksort working. It's not really quicksort though, since it's O(n^3) when the seq is already sorted.
15:25	hyPiRion	But if we assume rest is constant time, it's quicksort :)
15:28	TimMc	:-P
15:28	DerGuteMoritz	hello, I think I found a bug in clojurescript's core.cljs which requires a one-line fix. I haven't got a CA, yet, nor do I intend to get one right now. does anyone here care about it and wants to submit it as their own patch perhaps?
15:28	lynaghk	DerGuteMoritz: what is it?
15:29	DerGuteMoritz	lynaghk: I think the if-not in https://github.com/clojure/clojurescript/blob/master/src/cljs/cljs/core.cljs#L4428 is missing an else branch
15:29	DerGuteMoritz	with a (recur (inc i) init) call
15:30	DerGuteMoritz	I may be mistaken though!
15:31	DerGuteMoritz	hm, maybe I should take this opportunity and do the CA dance once and for all
15:31	lynaghk	DerGuteMoritz: How did you come across this? If you're just reading core.cljs for fun, then yeah, you should totally get a CA = )
15:32	hyPiRion	^ +1
15:32	DerGuteMoritz	lynaghk: I am actually porting PersistentHashMap to Scheme right now hehehe
15:32	DerGuteMoritz	ok I shall, then
15:32	lynaghk	DerGuteMoritz: ah. Yeah, I don't know enough about the internals to say if that's a bug or if someone used if-not when they meant when-not
15:32	amalloy	DerGuteMoritz: i wouldn't be surprised if this code is correct, though i'm not familiar with the internals
15:33	amalloy	ie, it seems unlikely to me that there would be a nil node followed by a non-nil node
15:33	DerGuteMoritz	let's see
15:33	DerGuteMoritz	well
15:33	DerGuteMoritz	but the correct thing to return then would be init, then, wouldn't it?
15:33	hyPiRion	A repro case is usually sweet to have.
15:33	DerGuteMoritz	yeah, I agree, that would be best
15:34	DerGuteMoritz	I will follow the code path completely again, maybe I overlooked something
15:34	amalloy	good point
15:34	TimMc	It could be a masked bug, i.e., not actually demonstrable until some other code changes and reveals it.
15:36	DerGuteMoritz	TimMc: yeah might be!
15:41	DerGuteMoritz	amalloy: for the record: empircally it seems to be possible that an ArrayNode has gaps
15:41	TimMc	That expression really points to the need for a let-loop-if-let-if-let-if macro.
15:42	DerGuteMoritz	absolutely
15:42	DerGuteMoritz	so this CA means I assign the copyright of contributions to Rich?
15:44	amalloy	DerGuteMoritz: you give him joint copyright, which is basically permission to do whatever he wants with it
15:44	DerGuteMoritz	ah yes, I just read the text
15:45	hyPiRion	DerGuteMoritz: It's there so that you cannot submit a patch, get it accepted and sue afterwards if you have a patent on the thing you submitted. IIRC.
15:45	ln2	(defn rest-of-vector [a]
15:45	ln2	(subvec a 1))
15:45	hyPiRion	Though don't quote me on that.
15:45	ln2	8)
15:46	amalloy	hyPiRion: that is the main reason it exists, yes
15:46	TimMc	hyPiRion: Also for relicensing.
15:46	jsabeaudry	out of curiosity, why don't I need a CA for contributing to the linux kernel?
15:46	DerGuteMoritz	hmmm so does that mean we could lose clojure completely if Rich decides to go proprietary? :-)
15:46	TimMc	Nah.
15:47	amalloy	DerGuteMoritz: no, because he's already licensed existing versions of it
15:47	amalloy	you could lose access to his latest and greatest new ideas if he decides not to share them with you, but that's hardly unusual
15:47	TimMc	In short, it would create a fork.
15:47	hyPiRion	amalloy: You won't lose access, you wouldn't get them to begin with.
15:48	DerGuteMoritz	ah, so it can't be used to change the past, good :-)
15:48	ln2	Is there any alternative to his database that is open source?
15:48	ln2	^Datomic
15:48	ravster	hello all
15:48	amalloy	ln2: postgres
15:48	DerGuteMoritz	ok guys, I gotta leave for a bit, I'll get back to you later when I know more about the possible bug
15:48	DerGuteMoritz	thanks for your feedback so far!
15:48	amalloy	mongodb. millions of open-source alternative databases. i realize this doesn't answer the question you meant to ask, but it's kinda a silly question :P
15:48	ln2	amalloy: Is it possible to mimic the same ideas with postgres? An accumulating data structure with timestamps ect?
15:49	ln2	amalloy: I was under the assumption that Datomic was unique.
15:49	ln2	ravster: Hello. =)
15:49	amalloy	shrug everything is unique
15:49	dnolen	DerGuteMoritz: that's one for Michal Marcyk to look at, if you open an issue and assign it to him he will probably respond pretty quickly.
15:49	ln2	amalloy: I was under the assumption that Datomic had capabilities that other databases didn't have?
15:49	amalloy	if you've been programming for four days i'd save worrying about datomic's licensing for later
15:50	gtrak	ln2: you can run datomic on postgres if you like.. what do you mean by possible to mimic the same ideas?
15:50	hyPiRion	ln2: Well, he has explained how Datomic works, so it's possible to create an open version of it.
15:52	ln2	amalloy: Understandable. The reason I'm drawn to Clojure is because I agree with the basic concepts. Even with programming for two weeks I think it's reasonable to start working with databases. No point in working with databases that don't agree with the basic concepts of the langauge. =)
15:52	DerGuteMoritz	dnolen: I can't open an issue without having signed the CA though, right?
15:52	hyPiRion	DerGuteMoritz: You can.
15:52	dnolen	DerGuteMoritz: anyone can open an issue
15:52	DerGuteMoritz	ah, cool, then I misunderstood that part
15:53	ln2	hyPiRion: That is what I was getting at. I understand that Datomic ultimately uses other databases as raw data stores. But the way that it places and organizes that data is unique (time based, imutable, ect.).
15:53	DerGuteMoritz	then I shall do that, thansk!
15:53	DerGuteMoritz	afk now for real :-)
15:53	hyPiRion	ln2: It unique in this moment, yes :)
15:54	ln2	gtrak: Yes. More along the lines of "Have those ideas been copied yet". It looks like the answer is no unfortunately.
15:55	gtrak	'a particular set of design tradeoffs', anything can do anything if you try hard enough
15:56	ln2	I wish I was a more competent programmer. I would help implement FreeTomic. ; )
15:57	ln2	It's unfortunate that Rich didn't kickstart Datomic, open source it and profit from support and solutions.
15:58	ln2	hyPiRion: Is there something missing from this or will this actually rest any vector?
15:58	ln2	(defn rest-of-vector [a]
15:58	ln2	(subvec a 1))
16:01	hyPiRion	ln2: It will indeed rest any vector.
16:01	ln2	hyPiRion: Sweet! I ran a bunch of tests but I wasn't sure if I had missed something stupid / obvious.
16:01	hyPiRion	Well, it won't return a seq, but it will return a collection without the first element
16:02	hyPiRion	oh right, we may have an odditiy.
16:02	hyPiRion	,(subvec [] 1)
16:02	clojurebot	#<IndexOutOfBoundsException java.lang.IndexOutOfBoundsException>
16:02	hyPiRion	Yeah, you need to handle that case.
16:02	ln2	An empty vector.....
16:02	TimMc	&(pop [])
16:02	lazybot	java.lang.IllegalStateException: Can't pop empty vector
16:02	hyPiRion	,(rest [])
16:02	clojurebot	()
16:03	ln2	& is an anonymous function?
16:03	lazybot	java.lang.RuntimeException: Unable to resolve symbol: is in this context
16:03	TimMc	ln2: No, & is lazybot's eval trigger.
16:03	ln2	Rest also fails against an empty vector? Interesting.
16:03	amalloy	hyPiRion: he only needs to handle that case if it's part of the function's contract, and he's making it up himself
16:03	TimMc	ln2: What do you mean by "fails"?
16:04	ln2	,(rest [])
16:04	clojurebot	()
16:04	hyPiRion	ln2: Doesn't fail, it returns ().
16:04	Raynes	That isn't failure.
16:04	Raynes	There is no rest of nothing.
16:04	ln2	I see it. Read the responses in the wrong order. =)
16:04	amalloy	there *ain't no rest of nothing
16:04	TimMc	heh
16:04	holo	hi
16:05	ln2	amalloy: Interesting. I'm trying to understand contracts. It really just means the "purpose" of the function as defined by me?
16:05	amalloy	yes
16:06	ln2	amalloy: If that purpose never involved empty vectors then theres no problem. I am still curious enough to try and fix that case.
16:07	gtrak	maybe being broken on empty it is part of the contract, conveniently, since that's what it does
16:07	ravster	cemerick: in 'friend', does the users data have to be a map of stuff that has username as a key and in the map itself?
16:08	ln2	gtrak: Maybe I just need to find a more productive problem to solve now that I mostly solved that one. Haha
16:09	holo	in korma, when a defentity users has-one email, this singular name "email" refers to a defentity email? or does it follow a plural names convention and it refers to defentity emails?
16:11	cemerick	ravster: maps are just convenient credential fns; see https://github.com/cemerick/friend/#credential-functions-and-authentication-maps
16:13	ln2	I tried a google on this one but can't seem to figure it out. What is the difference between the values 1 and :1?
16:13	ln2	Int vs Str?
16:13	jsabeaudry	Is validateur the defacto standard for parameter validation in a rest webapp ?
16:13	amalloy	jsabeaudry: no such standard has arisen yet
16:14	jsabeaudry	amalloy, what are the other top dogs?
16:15	amalloy	hell if i know. i just validate stuff by hand
16:15	Raynes	amalloy, jsabeaudry: lib-noir has a tiny validation thingy that I've never used and have nothing to do with.
16:17	ravster	cemerick: thanks
16:19	ppppaul	weavejester, you here?
16:20	SurlyFrog	Has anyone ever gotten a `ClassNotFoundException org.apache.http.impl.conn.PoolingClientConnectionManager` when trying to require clj-http.client ?
16:20	craigbro	nope
16:20	craigbro	however, if it was working previously...
16:20	amalloy	i'm guessing at least one person has
16:20	craigbro	sometimes rming target helps clean up confusion about deps
16:21	SurlyFrog	I'll try that
16:21	craigbro	gets me alot when switching between branches of a project that has different deps on different branches
16:21	holo	jsabeaudry, i think i lost several days trying and making market study with many of the validation solutions, only to end up rolling my own validation module. I still use the lib-noir email validation e.g., but i don't use it's stateful errors
16:22	weavejester	ppppaul: Yep
16:22	callenbot	Raynes: we need to talk.
16:22	ppppaul	what would be the easiest way to deal with having singular and pluralized routes map to the same fn?
16:22	callenbot	Raynes: but first: http://www.youtube.com/watch?v=7tOo2OMUhB8
16:23	weavejester	ppppaul: What do you mean?
16:23	callenbot	ppppaul: what like "/page/:id" and "/pages/:id" ?
16:23	ppppaul	yup
16:23	callenbot	weavejester: exercise a little imagination mate :)
16:23	weavejester	callenbot: Yes, but why would that be a problem?
16:24	Raynes	callenbot: Talk
16:24	callenbot	Raynes: what's the deal with signed cookies in lib noir?
16:24	Raynes	callenbot: EH?
16:24	SurlyFrog	removing target didn't help
16:24	jsabeaudry	holo, Interesting, what made you reject the libraries you found?
16:24	callenbot	Raynes: you can see the values.
16:24	callenbot	Raynes: it's key and key__s or whatever
16:24	gtrak	validateur seems to me no better than apply
16:24	callenbot	Raynes: and it just nukes the k/v pair if the __s doesn't match the value
16:25	Raynes	callenbot: I didn't write that, so shrug.
16:25	Raynes	I guess ibdknox likes broken shit?
16:25	callenbot	Raynes: balls. Enjoy the Misty Mountain Cold anyway. who do I harass?
16:25	callenbot	oh this is his fault?
16:25	callenbot	I'd better write a patch.
16:25	Raynes	Well, ibdknox wrote it but he won't be able to fix it.
16:25	callenbot	I know, that's why I realized I have to write it
16:26	Raynes	I'd take a patch to fix it, but I don't have time to do so myself anytime soon, especially since I don't know squat about cookies.
16:26	gtrak	"(numericality-of :age :even true)" I mean seriously? how about (even? x)
16:26	callenbot	Raynes: they're a sticking point / bugbear of mine. I'll fix it.
16:26	weavejester	ppppaul: Could you explain why plural vs. singular would be a problem?
16:26	weavejester	And for that matter, why you'd want it.
16:26	callenbot	weavejester: legacy URL compatibility?
16:26	ppppaul	it's not a problem. i'm just wondering how i would deal with this in as little code as possible
16:27	callenbot	weavejester: unless there's a legitimate x-y problem going on, it's usually a waste of time to pushback on peoples' stated needs. Sometimes they don't have a choice.
16:27	weavejester	ppppaul: Well, there's https://github.com/r0man/inflections-clj
16:27	ppppaul	legacy URL compatability is actually what i would like to deal with
16:27	weavejester	callenbot: Getting more information about the cause of the issue is almost never a bad idea.
16:28	callenbot	weavejester: maybe so but I've had a lot of bad experiences with time wasted questioning needs and having to re-explain an irrelevant use-case over and over.
16:28	ppppaul	in many cases in my app singular and plural don't have meaning for the api
16:28	holo	jsabeaudry, some were hard to understand, others did too much, and all of them (obviously) assumed i had to specify what kind of validation to run against each field. because i had to change constraints in the db, and repeat the process in the code, i built my own constraint map extracted from the db
16:29	rasmusto	hyPiRion: thanks for the leiningen patch (`lein clean`)
16:30	technomancy	ln2: for the record, clojars uses a data store that has the same assumptions of immutability and append-only as datomic, but since the data set is small (and not distributed) it's much easier to implement.
16:30	technomancy	oops
16:31	weavejester	callenbot: I've mostly had the opposite problem. Asking one or two questions before leaping to answer a problem is often a sound idea, and usually doesn't take up much time.
16:31	holo	jsabeaudry, i also use conventions of my own extracted from the database (but not constraints) to validate for errors that are not available from constraints. e.g. validate for email when a field's name has email in the name
16:31	ppppaul	thanks for the link weavejester, however i'm talking about matching routes. do i have to duplicate my routes then? clout can't handle something like what i'm asking for?
16:32	callenbot	weavejester: fair. also if I'm first person to notice the sucky cookies implementation then how have you been doing auth/saved-signin all this time?
16:32	gfredericks	what kind of storage does datomic free use? some opaque internal thing?
16:33	hiredman	gfredericks: looks like hsql
16:33	hiredman	gfredericks: but that is just a guess since I'g gotten hsql exceptions from broken datoic free setups
16:33	desertmonad	so is it "day-tom-ic" or "duh-tom-ic"?
16:33	weavejester	ppppaul: Hum. You could look for directory paths in the path and setup some regular expression to match both parts.
16:33	weavejester	callenbot: Sucky cookies implementation?
16:34	weavejester	callenbot: What do you mean?
16:34	weavejester	ppppaul: Or… maybe a "normalize route" middleware.
16:35	weavejester	ppppaul: Split a route up by /, then pluralize any singular words.
16:35	gfredericks	hiredman: cool thanks
16:35	callenbot	weavejester: signed cookies in lib-noir upsets me :)
16:35	gfredericks	desertmonad: this is english, so yes
16:36	jsabeaudry	holo, I see, do you remember the names of the better ones you considered? (besides validateur and lib-noir)
16:36	weavejester	callenbot: I hadn't seen that before. I don't have a lot of interation with lib-noir.
16:37	weavejester	What's the purpose of put-signed?
16:37	callenbot	weavejester: okay fine but that wasn't my question
16:37	callenbot	weavejester: I was asking you how you kept users logged in
16:37	weavejester	callenbot: With cookies? I'm not sure I understand the question.
16:37	callenbot	weavejester: you do use cookies?
16:38	weavejester	callenbot: Yes, but not lib-noir.
16:38	callenbot	weavejester: what do you save, an id or something? How do you keep them from tampering with the cookie?
16:38	callenbot	are you signing them yourself?
16:39	weavejester	callenbot: It depends. Ring has several different cookie stores. The most common mechanism is to generate a UUID and link that to a database value that contains session data.
16:39	amalloy	callenbot: everyone just sends the user an id, and keeps the real data server-side somewhere
16:39	callenbot	amalloy: a db-backed session id?
16:39	weavejester	callenbot: But Ring also has a cookie-store implementation that encrypts the session data with AES and HMACs it with a secret key.
16:39	ppppaul	weavejester, i think i'll use multiple contexts with inflections-clj that point to the same route
16:39	callenbot	weavejester: wait wait, where's this cookie-store implementation?
16:39	holo	jsabeaudry, i can't precise it now after so much time, but it's safer to go to clojuresphere.com and check how many projects are using foo and bar validation libs. if no one is using it, suspect something is wrong with the lib :)
16:40	callenbot	weavejester: that's what I need. is something that hides the kv pairs.
16:40	weavejester	callenbot: https://github.com/ring-clojure/ring/blob/master/ring-core/src/ring/middleware/session/cookie.clj
16:40	callenbot	I don't like db-backed session stores.
16:40	weavejester	callenbot: Also read: https://github.com/ring-clojure/ring/wiki/Sessions
16:40	weavejester	I believe lib-noir has a stateful session implementation based on top of Ring's sessions.
16:41	hyPiRion	rasmusto: Oh, you're welcome :)
16:42	jsabeaudry	holo, oh wow thanks, I didn't know about clojuresphere
16:42	technomancy	hyPiRion has been on fire recently =)
16:42	weavejester	callenbot: (wrap-session handler {:store (cookie-store secret-key)})
16:42	callenbot	weavejester: yeah I see it. I'm pondering how to make the wrapping work with my template rendering
16:42	holo	jsabeaudry, you're welcome
16:42	callenbot	weavejester: thanks for the pointer, it's much appreciated.
16:43	technomancy	http://www.generationgamerz.com/images/post/361/nbaj_02.jpg
16:43	rasmusto	technomancy: that symlink? code seems like it should exist somewhere else, are symlinks just not used frequently in clojure projects? (my case may have been special)
16:43	weavejester	callenbot: No problem. You can take the session data out and put it into a binding, of course. And lib-noir provides stateful session functions.
16:43	callenbot	technomancy: damn dude, now I want to play that.
16:44	gtrak	technomancy: sb3j3c2 nyhkhbs is a cheat-code for some reason I can never forget. haven't played in 13 years.
16:44	holo	jsabeaudry, i would gladly provide you my module, but unfortunately it has too much code smell, and is coupled with database details
16:44	weavejester	So you could write: (wrap-noir-session handler {:store (cookie-store …)})
16:44	callenbot	weavejester: I don't like the stateful session stuff in lib-noir. I have my own cookie and request context stuff floating around
16:44	weavejester	And then (noir.session/get "foo")
16:44	weavejester	(noir.session/put! "foo" "bar")
16:45	technomancy	rasmusto: it'd be nice if it were in clojure.java.io, but it's small enough that it's probably not worth the effort that would take
16:45	weavejester	Behind the scenes it uses an atom that exists for the duration of the request.
16:45	weavejester	I usually prefer a more functional approach, however.
16:46	TimMc	hyPiRion: Nice solution.
16:46	callenbot	weavejester: I have my own atom thing I'm using that does that.
16:46	hyPiRion	TimMc: Chinese wall of Apache Commons solution, praise them.
16:46	holo	weavejester, are you able to sanely not use session state?
16:47	rasmusto	technomancy: understood
16:48	TimMc	hyPiRion: I was trying to work something like that out yesterday. I didn't think of taking the canonical parent and sticking the last path element back on.
16:48	weavejester	holo: I just use session state for holding the user ID.
16:48	weavejester	holo: I've never needed anything more.
16:49	callenbot	weavejester: I do the same thing.
16:49	callenbot	weavejester: I do need a RequestContext thingamajig to keep from getting pissed off though.
16:49	hyPiRion	TimMc: Well, now you know that trick!
16:49	gtrak	how do you expire a cookiestore session without another store?
16:49	weavejester	callenbot: Did you implement that?
16:50	weavejester	gtrak: Cookies have their own expiration mechanism.
16:50	holo	weavejester, amazing! after some unsuccessful refactorings, i came with the same conclusion (yesterday) - just store the user ID
16:50	hyPiRion	To be fair, I was hoping for a cleaner solution, but the "Java on all platforms" kind of breaks down when you're dealing with system specific things.
16:50	gtrak	ah, tell the browser to expire it and hope it complies
16:51	weavejester	gtrak: Well, if you want to be certain about it, you can set an internal timestamp and check it.
16:51	TimMc	hyPiRion: "Runs on all platforms" is like "it's Turing complete!".
16:52	callenbot	weavejester: yes but it's not an original idea. I based it on Django/Flask RequestContext and data lifecycles
16:52	gtrak	ah
16:52	holo	weavejester, ok, that's a lie. i actually store the username too, but the ID is the most important since it's reliable cause it never changes
16:52	weavejester	holo: I can't really think of anything else you'd want to store. People give examples like shopping baskets and such, but they're usually better implemented as resources on the server.
16:52	callenbot	weavejester: I implemented it myself because none of the people who hadn't used Python before had any idea what the hell I was talking about.
16:53	weavejester	callenbot: I'd encourage you to release it as middleware, if you have time.
16:53	callenbot	weavejester: you people will hiss at me because it's not functional.
16:54	rbarraud	You people?! <ninja_pose> (Me, myself and Irene) :-/
16:54	holo	weavester, i store the username, cause every page salutes the user >.> .. i thought it would be slugish to make a query on every page change hehe.. and it's not a big deal if the username is not up to date with the db for this purpose
16:54	rbarraud	Rofl
16:54	technomancy	we wouldn't do that
16:54	technomancy	we have bots to do the hissing for us now
16:54	TimMc	$hiss
16:54	TimMc	:-(
16:54	rasmusto	clojurebot: hiss
16:54	clojurebot	Gabh mo leithscéal?
16:54	callenbot	technomancy: :D (btw, I was talking to me best friend who also moved to Cali about Seattle. He might follow me there if I move)
16:54	TimMc	clojurebot: boo is <reply>hiss
16:54	clojurebot	In Ordnung
16:55	weavejester	callenbot: Sometimes one has to depart from pure functional code, but pure functions are very nice to work with.
16:55	rbarraud	Clojurebot: rant
16:55	weavejester	holo: Caching the username seems reasonable.
16:55	callenbot	weavejester: alright I'll do it. Do you use lib-noir?
16:55	rbarraud	Fail...
16:55	weavejester	callenbot: Nope.
16:56	callenbot	weavejester: you said you didn't right? okay.
16:56	callenbot	I'll try to keep it ring-only then.
16:56	TimMc	clojurebot: rant is http://www.ccs.neu.edu/home/shivers/rants/
16:56	clojurebot	You don't have to tell me twice.
16:56	rasmusto	clojurebot: rant
16:56	clojurebot	rant is http://www.ccs.neu.edu/home/shivers/rants/
16:56	weavejester	callenbot: I'm not advocating avoiding lib-noir. If it's useful, by all means, use it :)
16:57	weavejester	But lib-noir tends to advocate a slightly less functional approach than Ring.
16:57	callenbot	weavejester: I'm just saying that I'll strive for reusability and composability.
16:57	weavejester	callenbot: I'm all for that :)
16:57	abp	callenbot: That's need a (def (atom ..)) per namespace!
16:57	weavejester	Which reminds me...
16:57	rbarraud	clojurebot: rant
16:57	clojurebot	rant is http://www.ccs.neu.edu/home/shivers/rants/
16:58	weavejester	I need to release crypto-password
16:58	callenbot	abp: what?
16:59	abp	yea the sentence was f**d up
16:59	callenbot	abp: wat
17:00	abp	callenbot: For reusability and composability you need at least one (def x (atom ...)) per namespace. :P
17:00	gtrak	I thought the point of dynamic binding was that it wasn't composable?
17:00	callenbot	abp: no I don't
17:00	callenbot	abp: my thingy works fine.
17:01	abp	I were joking.
17:01	callenbot	I use it to shoot stuff from pre-processing middleware to the templates.
17:01	callenbot	abp: I don't have a sense of humor.
17:02	abp	And I shouldn't do jokes based on my opinions in a technical irc-channel, probably.
17:02	holo	callenbot, i find the defpages from noir really sweet
17:02	callenbot	holo: I don't do webapps that way.
17:02	callenbot	holo: most of my web work is collaborative/team based
17:03	callenbot	aka I'm working with a frontend guy...that doesn't know Clojure.
17:03	gtrak	Vars shouldn't be in libraries. I hate it when people make assumptions about lifecycles for me.
17:03	abp	So no hiccup for you?
17:03	callenbot	gtrak: so don't use it. boohoo.
17:04	callenbot	gtrak: I need it and so do a lot of web apps.
17:04	hyPiRion	callenbot: The obvious solution is to learn them Clojure.
17:04	callenbot	gtrak: it's just ring middleware.
17:04	callenbot	hyPiRion: we don't want to repeat this conversation.
17:04	callenbot	hyPiRion: I've been working and running teams long enough to know what's in the cards and what ain't
17:04	gtrak	callenbot: just saying, vars are more useful for an application... and not talking about you specifically, other libs have the same problem
17:04	callenbot	hyPiRion: guess what? it ain't in the fucking cards.
17:05	callenbot	gtrak: well I need a way to shoot data around the handlers. so blah.
17:05	abp	callenbot: That can't be put into the request-map?
17:05	gtrak	yes, I actually like vars more than most of my colleagues, just at the top
17:05	weavejester	gtrak: You mean like with-connection or something?
17:06	gtrak	weavejester: I think so?
17:06	callenbot	abp: hrm. I just got an idea.
17:07	TimMc	&(format "%02x" (int \#))
17:07	lazybot	⇒ "23"
17:07	gtrak	weavejester: yea. I prefer to write extra glue than do surgery on a namespace
17:07	brianwong	(for [x (range 4)] (println x))
17:07	abp	callenbot: Around getting rid of state? :D
17:07	brianwong	is the reason the nils are at the end in REPL because of the lazy evaluation?
17:07	abp	mutable*
17:08	weavejester	gtrak: Agreed. I don't like side channels, especially when they're not written by me ;)
17:08	holo	any good advice for a unix shell script lisp? or i will just try fish. it looks nice
17:08	brianwong	as in the side-effects are not guaranteed to happen before the next item is returned?
17:08	gtrak	I really like vars that I write :-)
17:08	gfredericks	brianwong: yeah
17:08	gfredericks	brianwong: the side effects happen as the repl walks through the lazy seq to print the nils
17:09	gfredericks	brianwong: wrap the thing in doall to see a different effect
17:12	ubuntu1	hi
17:12	gfredericks	hi.
17:12	holo	hi
17:12	brianwong	gfredericks: why doesnt the (for [x (range 4)] (println x)) print anything?
17:13	hiredman	~for
17:13	clojurebot	for is complected
17:13	hiredman	~for
17:13	clojurebot	for is not used often enough.
17:13	hiredman	~for
17:13	clojurebot	for is not a loop
17:13	nDuff	brianwong: for creates a lazy sequence.
17:13	brianwong	a list comprehension
17:13	gfredericks	hiredman: oh oh is there inference on "is not"? :D
17:14	nDuff	brianwong: ...if you want it to execute immediately, and you aren't looking for side effects, you should probably be using doseq instead.
17:14	gfredericks	nDuff: s/aren't/are/?
17:14	nDuff	err, right.
17:14	brianwong	i see
17:15	gtrak	sequences are about values, side effects aren't
17:15	gfredericks	laziness and side effects aren't comfortable together
17:15	gfredericks	they fight a lot
17:15	hiredman	gfredericks: I think the verb there is \|is\| so you would have to say "not a loop is xyz"
17:15	brianwong	(doc doall)
17:15	clojurebot	"([coll] [n coll]); When lazy sequences are produced via functions that have side effects, any effects other than those needed to produce the first element in the seq do not occur until the seq is consumed. doall can be used to force any effects. Walks through the successive nexts of the seq, retains the head and returns it, thus causing the entire seq to reside in memory at one time."
17:16	brianwong	it says that the side effects of a lazy sequence would happen after the sequence is consumed
17:16	gfredericks	the "other than those needed to produce the first element" sounds wrong?
17:16	brianwong	so shouldnt (for [x (range 4)] (println x)) produce the side-effects?
17:16	nDuff	brianwong: What's consuming it?
17:16	brianwong	even at the end?
17:16	brianwong	ahhh
17:16	brianwong	nDuff: i see your point
17:16	brianwong	nothing is using the lazy sequence
17:17	nDuff	Exactly.
17:17	nDuff	brianwong: (btw, it's generally "during" rather than "after")
17:17	nDuff	...and actually, can even be before with chunked sequences
17:19	hyPiRion	The only promise is that the side effects will be done before the element is realized.
17:19	gtrak	brianwong: also, seqs are read in chunks ##(first (for [x (range 4)] (println x)))
17:19	lazybot	⇒ 0 1 2 3 nil
17:20	gtrak	notice first makes all four happen
17:20	gfredericks	gtrak: only chunked seqs
17:20	gtrak	gfredericks: sure :-), the point is to foster a healthy paranoia
17:20	gfredericks	phew
17:21	brianwong	interested
17:21	brianwong	interesting
17:22	brianwong	thanks for the help guys
17:22	gtrak	##(first (for [x (range 50)] (println x)))
17:22	lazybot	⇒ 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 nil
17:23	gfredericks	por eso estamos
17:30	thalassios_xelon	heloo guys:)))
17:30	thalassios_xelon	,(+ 1 2)
17:30	clojurebot	3
17:30	holo	maybe some girls are present too
17:31	thalassios_xelon	girls are not clever :P
17:31	thalassios_xelon	hello holo :)
17:31	holo	hi!
17:31	thalassios_xelon	holo u girl?
17:31	nDuff	...eh, easier to /ignore
17:31	brianwong	hahah
17:32	holo	i'm a male
17:32	gfredericks	don't treed the folls
17:33	thalassios_xelon	holo are you blond and sexy clojure geek?
17:33	holo	>.>"
17:33	tmciver	thalassios_xelon: I'm a blond and sexy geek . . . but I'm a guy. ;)
17:33	thalassios_xelon	bye :*
17:34	gtrak	(votekick thalassios_xelon)
17:34	thalassios_xelon	gtrak, dont be evil
17:34	gfredericks	(dec thalassios_xelon)
17:34	lazybot	⇒ -1
17:34	TimMc	C'mon, don't feed. Just /ignore.
17:34	gfredericks	TimMc: good point. done.
17:34	nDuff	(inc TimMc)
17:34	lazybot	⇒ 34
17:35	holo	(inc holo)
17:35	TimMc	(And if you're an op, kick.)
17:35	lazybot	You can't adjust your own karma.
17:36	TimMc	gtrak: I think I've heard of IRC clients where /ignore is broadcast/made available to other clients.
17:36	arohner	given defrecord foo, is there an existing way to find its (map->Foo) fn?
17:36	brianwong	(brianwong)
17:36	TimMc	A quorum makes compliant clients auto-ignore.
17:36	gfredericks	arohner: find it?
17:36	technomancy	inc
17:36	ivan	)/whois thalassios_xelon
17:36	brianwong	##(inc 1)
17:36	lazybot	⇒ 2
17:36	gtrak	TimMc: ah, fun, it's like git vs svn
17:36	brianwong	##brianwong
17:36	amalloy	TimMc: i would want that so hard
17:36	brianwong	i guess i have no value
17:37	brianwong	that makes me sad
17:37	xeqi	$karma brianwong
17:37	lazybot	brianwong has karma 0.
17:37	arohner	gfredericks: I want a fn like (resolve-map-fn Foo) that returns the result of (resolve 'map->Foo)
17:37	gtrak	can we build a git repo for /ignores and have a web of trust?
17:37	arohner	but I want to pass arbitrary records in
17:37	arohner	it looks possible using (.getName record) and (.getSimpleName record), but that's a total hack
17:37	gfredericks	arohner: you want to pass in an instance? or the class?
17:37	arohner	the class
17:37	sjl	Do Leiningen and Drip work together yet? I followed the instructions but I'm not really seeing any speed boost
17:37	holo	$karma holo
17:37	lazybot	holo has karma 0.
17:38	nDuff	gtrak: Hmm. There'd have to be standards established in terms of tolerance. I've got a lot of entries on my /ignore list from time-wasters as opposed to outright trolls, for instance, but others might not want that.
17:38	gfredericks	arohner: yeah I'm not sure there's anything for that beyond what you're doing
17:38	gfredericks	nDuff: we could say /troll
17:38	TimMc	nDuff: Another problem is that ignores make you blind to a problem, they don't remove it.
17:38	gtrak	nDuff: easily solved with an int rating
17:38	aaelony	hyPiRion: fyi, figured out the command i needed earlier for profiling. java -agentpath:/path/to/YourKit_Java_Profiler_12.0.2.app/bin/mac/libyjpagent.jnilib -jar my-uberjar.jar
17:39	technomancy	sjl: they're supposed to, but I've seen reports of non-workage. are you on lein 2.0.0 final?
17:39	ivan	jcowan/Chouser/rhickey can add people to the access list
17:39	ravster	what would a credential-fn look like for the http-basic workflow in 'friend'?
17:39	thalassios_xelon	good night room :)
17:40	sjl	technomancy: http://paste.stevelosh.com/51142d5509b5f60007000001
17:40	sjl	I think that means 2.0.0 final?
17:40	sjl	just says 2.0.0
17:40	technomancy	sjl: yeah, that's the latest
17:40	amalloy	and 0.2.3 is latest drip, i'm fairly sure
17:40	technomancy	sjl: sorry; haven't used drip myself, but I saw some patches go in right before lein 2.0.0 landed.
17:40	sjl	it does seem to start the extra JVM: http://paste.stevelosh.com/51142d2109b5f60007000000
17:41	hyPiRion	aaelony: oh, sweet.
17:41	arohner	gfredericks: I just wrote: https://gist.github.com/arohner/4734884
17:41	sjl	doesn't really speed up much though
17:41	aaelony	:)
17:41	amalloy	sjl: looks like https://github.com/flatland/drip/issues/57
17:42	amalloy	ninjudd is out of the country for the week, and i'm not really qualified to fix anything, so this will probably get fixed less quickly than usual
17:42	sjl	so the answer is no, they still don't work together
17:42	sjl	hah
17:44	technomancy	there is also https://github.com/technomancy/leiningen/issues/968
17:44	technomancy	which doesn't exist yet
17:44	technomancy	actually I think I have a stash for that; I should at least put it on a branch
17:45	amalloy	sjl: what if you time lein trampoline test?
17:45	amalloy	i'm thinking that lein test should be using two JVMs (right, technomancy?), and only one of them is being controlled by drip
17:45	technomancy	amalloy: yeah
17:46	sjl	amalloy: same thing -- user time drops but real time is (roughly) the same
17:46	technomancy	sjl: `export LEIN_FAST_TRAMPOLINE=y` should allow trampoline calls to avoid launching Leiningen's own JVM entirely for trampolined invocations.
17:46	gfredericks	arohner: coo
17:46	jweiss	is there a better way to split a list of ints at the max value, other than using map-indexed (or otherwise producing a structure that contains indices)?
17:46	technomancy	there's also my eshell launcher that avoids the bash script entirely and uses nrepl
17:46	technomancy	which doesn't work either
17:47	hyPiRion	jweiss: Is there only one max value?
17:47	technomancy	anyway, the point is there are a bunch of ways to do it and none of them work all that well, in conclusion yay open source
17:47	amalloy	jweiss: i don't think that's possible lazily
17:47	jweiss	hyPiRion: no, there could be more than 1, but it should be split at the first.
17:47	sjl	technomancy: no difference
17:47	technomancy	sjl: sorry, I meant repeated calls should be faster
17:48	technomancy	with LEIN_FAST_TRAMPOLINE
17:48	sjl	oh
17:48	sjl	yes they are, a bit
17:48	sjl	like 1s instead of 3
17:48	technomancy	repeated calls to the exact same command anyway
17:48	sjl	yeah
17:48	sjl	technomancy: what's the downside to setting this var?
17:48	technomancy	it memoizes the resulting form to call from that task
17:48	technomancy	sjl: it's not well tested
17:48	sjl	ah
17:48	technomancy	at some point it should default to on though
17:48	hyPiRion	,(let [orig (repeatedly 6 #(rand-int 4000)), m (reduce max orig)] [orig (split-with (partial not= m) orig)])
17:48	clojurebot	[(3109 1760 1057 2537 1965 ...) [(3109 1760 1057 2537 1965) (3144)]]
17:49	jweiss	hyPiRion: thanks!
17:49	technomancy	sjl: there are probably some cache invalidation issues with it where the memoized command would end up being out of date
17:49	hyPiRion	jweiss: Not sure if it's better, per se.
17:49	sjl	technomancy: makes sense
17:49	hyPiRion	But it's at least without indices
17:49	technomancy	if that happens; `lein clean` will take care of it
17:49	jweiss	hyPiRion: i think that traverses the list twice though.
17:49	technomancy	but changes to project.clj should get picked up
17:49	sjl	cool
17:49	technomancy	but (for instance) new versions of a snapshot won't
17:49	amalloy	jweiss: you can't help but traverse the list twice
17:50	technomancy	because project.clj doesn't change, but the proper command does
17:50	hyPiRion	yeah, you have to find the max.
17:50	technomancy	Fraught with Peril™
17:50	jweiss	amalloy: can't you split a vector at an index without traversing it?
17:51	jweiss	find the max should be the only traversal
17:52	hyPiRion	jweiss: is it a vector we're talking about?
17:52	amalloy	unless your first traversal is to put it into something random-accessible
17:52	amalloy	eg, you could reduce over the sequence, putting it into a vector and finding the max element; then split up the resulting vector with two subvec calls
17:52	amalloy	actually, i tell a lie. of course you can do this in one pass
17:52	amalloy	jweiss: that's what i just said. if you put it into something random-accessible
17:53	jweiss	ok thanks
17:54	gfredericks	twiceish
17:55	gfredericks	use split-at and do it in 3 traversals!
17:56	amalloy	jweiss: https://www.refheap.com/paste/abcc0036c84dc85622b0f69ce is one way that's interesting, if not necessarily good
17:56	gfredericks	I guess split-with has that property too
17:56	hyPiRion	&(first (reduce-kv (fn [[i max :as init] j v] (if (< max v) [j v] init)) [-1 -1] [1 2 10 3 4]))
17:56	lazybot	⇒ 2
17:56	hyPiRion	then do subvec from there.
17:58	amalloy	but really in a lazy language thinking about the number of times you traverse something is very confusing
17:58	dnolen	unifier in core.logic is getting fancy - http://gist.github.com/swannodette/4735012
17:58	gfredericks	we need a contest where we try to maximize CPU-cycles / source-char
17:59	gfredericks	it'd be like clojure golf but with an anti-performance component
17:59	amalloy	gfredericks: that'd be a short contest
17:59	gfredericks	amalloy: sleeping doesn't count
17:59	amalloy	(#(recur))
17:59	gfredericks	hrm
17:59	gfredericks	amalloy: also you have to solve the problem at hand
17:59	gfredericks	like splitting a list at the max element
18:00	ppppaul	i have a destructuring issue that i would like help with. https://gist.github.com/boxxxie/4735021 :D
18:00	gtrak	(dotimes [x 1000] (future (recur)))
18:00	gtrak	999, rather
18:01	gfredericks	ppppaul: ah ha
18:01	gfredericks	ppppaul: so line 7
18:01	gfredericks	wait are you trying to make something equilavent to line 2?
18:01	ppppaul	gfredericks, yes
18:02	gfredericks	ppppaul: okay remove the curly braces around {:keys [body content-length]}
18:02	amalloy	gfredericks: you might find http://en.wikipedia.org/wiki/Busy_beaver entertaining, then
18:02	gfredericks	amalloy: man I gave a lightning talk on that
18:02	gfredericks	I don't think I did it well though
18:03	ppppaul	gfredericks, can you post on my gist?
18:03	gfredericks	specifically it was about the "busy beaver sequences grows faster than any computable sequence" part
18:03	gfredericks	ppppaul: yeah
18:03	dnolen	lynaghk: pretty much done w/ vector & set syntax
18:04	ppppaul	thanks gfredericks
18:04	gfredericks	np
18:06	amalloy	i apparently don't even know what noncomputable really means
18:06	lynaghk	dnolen: awesome!
18:07	gfredericks	amalloy: is it mentioned in the article?
18:07	lynaghk	dnolen: I've been in meetings and haven't had a chance to look yet =/
18:07	amalloy	gfredericks: yes
18:07	gfredericks	amalloy: any sequence of numbers you can write a function to compute will provably grow slower than the busy beaver sequence
18:07	dnolen	lynaghk: the biggest change is that unifier only really takes 1 or 2 args now. in 1 arg case no options - just sequences of terms to unify. 2 args case first arg is the options.
18:08	dnolen	lynaghk: set & vector syntax both support giving a single constraint, or a seq of constraints
18:08	gfredericks	amalloy: the proof is a one-line reduction to the halting problem
18:08	dnolen	lynaghk: there are tests for these different cases - let me know if they don't make sense.
18:09	lynaghk	dnolen: okay, will do.
18:22	gfredericks	json with cheshire should be a faster serialization than pr-str/read-edn, right?
18:23	hiredman	gfredericks: I don't think there is an inherent reason it should be so, but I believe it is
18:23	dxeh	_veer,
18:23	dxeh	super_?
18:23	clojurebot	contribute is http://clojure.org/contributing
18:23	hiredman	e.g. someone should write a high performance edn codec
18:24	gfredericks	I wonder if rhickey still wants tools.edn since he added read-edn?
18:24	_veer	dxeh: ?
18:24	mrb_bk	dnolen: oooh nice unifying
18:24	dxeh	do you have another alias "super_"
18:24	technomancy	sjl: highly experimental: https://github.com/technomancy/leiningen/issues/968
18:25	dnolen	mrb_bk: yes, should make the unifier a lot more useful.
18:25	_veer	dxeh: why?
18:25	dxeh	cause if you do then maybe you know me _veer :P
18:26	_veer	okay
18:29	mrb_bk	dnolen: we should grab a beer and talk about how i can help with some documentation stuff soon
18:30	dnolen	mrb_bk: hey, that sounds like a great idea!
18:31	mrb_bk	dnolen: awesome, I'll drop you an email
18:32	mrb_bk	i made my first proper minikanren goal in core.logic and almost jumped out the window in excitement
18:32	dnolen	mrb_bk: haha, it's pretty great when it clicks.
18:35	technomancy	wow, vimclojure ships with vim now
18:36	technomancy	that should help a lot considering installation has been somewhat involved from what I hear
18:36	mrb_bk	technomancy: confirmed
18:37	technomancy	I guess that means vim doesn't have a CA policy; good for them =)
18:37	technomancy	clojure-mode will never ship with Emacs unfortunately due to that
18:40	bawr	CA?
18:40	clojurebot	CA is Contributor Agreement: http://clojure.org/contributing
18:41	bawr	Oh, that's useful.
18:50	rasmusto	if I turn a map into a seq, what's the good way to turn that back into a map? I'm mapping a function to my map, returning a seq of k/v vectors, then wanting to turn that into a map
18:50	amalloy	(into {} kvs)
18:51	rasmusto	amalloy: ok, thanks
18:56	DerGuteMoritz	ok I think I found a way to hit the potential bug
18:57	DerGuteMoritz	now I wonder how to test my patch
18:57	DerGuteMoritz	i.e. how do I run the cljs test suite?
18:58	DerGuteMoritz	does anyone know?
18:58	DerGuteMoritz	ah, script/test I guess
19:03	Raynes	In case anyone wonders why refheap suddenly has 11k pastes vs the 9k it used to have, it's because someone spammed the crap out of it recently, apparently. https://www.refheap.com/paste/11032 is the last 1000 or so pastes.
19:04	Raynes	And congratulations to whoever did this act of elite hackery, pat yourself on the back for spamming the most un-ratelimited pastebin in history.
19:05	technomancy	mad skillz
19:06	Raynes	Unless of course, as amalloy just pointed out, it could be unintentional. If that's the case, please let me know and complain about there not being a sandbox for testing API stuff so I can point you to the sandbox for testing API stuff.
19:07	technomancy	it doesn't look spammy spam
19:08	amalloy	yeah, it looks like someone is running https://github.com/killswitch/dunn/blob/master/plugins/git.js and has it configured to automatically refheap exceptions
19:09	shaunxcode	someone asked earlier if graph supported clojurescript - mid way down the readme: "Also, stay tuned for a ClojureScript version of Graph, coming soon."
19:09	Raynes	Hrm
19:09	amalloy	Raynes: https://github.com/killswitch/dunn/blob/master/libs/irc.js#L90
19:09	Raynes	I see.
19:09	Raynes	So this is a totally valid usage.
19:10	amalloy	well, valid usage run amok by him having too many errors. i'd get in touch with him and ask him to chill out
19:10	technomancy	depends on your TOS =)
19:10	Raynes	I mean, maybe not sending 1k pastes.
19:10	abp	technomancy: In Emacs, can I open the readme.md of a dependency of a Clojure-project I'm working on? M-. let's me jump to a symbols definition. But then from there, I can't further navigate in the jars context apparently.
19:10	Raynes	But I don't have any rules.
19:10	technomancy	abp: M-. into a file in a jar will open both that file and the jar itself
19:11	technomancy	so look in your buffer list
19:11	technomancy	abp: that said, we don't include readmes inside the jars right now. there's an open issue for it though.
19:12	abp	technomancy: Oh ok. Also the jar isn't in my buffer list. But this is on windows..
19:12	technomancy	shouldn't matter. are you sure you've opened a .clj file in a jar?
19:13	technomancy	if so C-x C-f would default to being deep inside .m2/repository
19:14	abp	technomancy: yes, buffer-list points into .m2/.. C-x C-f is in there too
19:14	abp	technomancy: But C-x C-f context being right is enough. I somehow got the impression thats not the case.
19:15	dxeh	clojure seems quite the interesting language, coming from c/java programming its quite the switchup which is nice
19:15	dxeh	syntax wise, that is
19:15	technomancy	huh... I guess that might vary by OS then; that is surprising to me.
19:16	abp	technomancy: Should the jar be opened like a dired buffer..?
19:16	technomancy	abp: yeah, that's how it works here
19:16	ravster	abp: jars in emacs open automatically for me.
19:16	technomancy	you can see all the .class files and stuff
19:16	ravster	I don't need to do anything special
19:19	abp	technomancy: I'm fine with it, I can walk into the jar from the files context, dired-like.
19:19	technomancy	ok, cool
19:20	technomancy	feel free to chime in here on the topic of including readmes: https://github.com/technomancy/leiningen/issues/880
19:21	ravster	I'm trying to make a credential-fn for 'friend' for the project I'm working on, but I'm getting 'wrong number of args passed [0]' errors. How do I make sure that the cred-fn gets the map of user and password thats coming in from the request? I'm using http-basic workflow.
19:21	abp	technomancy: Using emacs for two days now, preceded by a few days of reading the tutorial, memoizing basic bindings, reading .emacs.d's, now using emacs-live. I'm extremely impressed and addicted. :)
19:22	technomancy	cool; have fun =D
19:23	brehaut	waitaminute, you read the documentation‽ thats not doing it right!
19:24	technomancy	brehaut: you're supposed to write documentation, not read it.
19:24	brehaut	technomancy: wait, you sure you are in the right community?
19:25	dxeh	what is reading?
19:25	technomancy	brehaut: sorry, "you're supposed to chide people for not writing documentation, not for not reading it"
19:25	Raynes	You guys sound like antares all of a sudden.
19:25	technomancy	now I'm just confused
19:25	brehaut	technomancy: PHEW
19:26	tanzoniteblack	shaunxcode: the clojurescript comment was obviously way too unclear...I mean it's in its own paragraph in the readme, how could someone possibly notice it?
19:27	abp	Readme's being way to long with more then one sentence will be all the rage soon.
19:27	dxeh	is it true there is a "clojurescript" which compiles clojure to js rather than java lmao
19:28	technomancy	lol its true
19:28	shaunxcode	tanzoniteblack: hah I read the whole thing as it is absolutely solving a problem I face
19:29	tanzoniteblack	shaunxcode: I'm doing the same; less because it solves a problem I currently face, but more because it might be useful in the future and my brain hurts from debugging all day
19:29	tanzoniteblack	https://github.com/Prismatic/plumbing in case anyone is wondering what readme we're referring to
19:29	dxeh	technomancy, you wrote leiningen?
19:30	technomancy	dxeh: guilty as charged
19:30	craigbro	as transcribed to him by aiwas
19:30	craigbro	rather, whispered into his ear
19:30	dxeh	impressive technomancy
19:30	craigbro	just like Aleister Crowley
19:30	abp	shaunxcode: Something related to your datomic workspace?
19:31	dxeh	technomancy, i was just like lets look up some random people on github and i laughed so hard when i typed in your name and saw leiningen
19:34	amalloy	i think he's calling you random, technomancy
19:34	dxeh	the world is a random place
19:34	tommo	what does a : denote before something in clojure, e.g. [:a :b]
19:35	brehaut	tommo: keyword literal
19:35	tanzoniteblack	tommo: http://clojure.org/data_structures#Data%20Structures-Keywords
19:35	brehaut	(a keyword is a symbol that refers to itself rather than something else)
19:38	ravster	cemerick: how should I go about making a credential-fn for http-basic that hooks into a function that I already have that checks usernames & passwords against those in a DB?
19:38	shaunxcode	abp: yes actually, I have actually started working on feature enhancements in the coffeescript branch whilst I wait til I am happy with "best practice" in cljs branch
19:39	cemerick	ravster: you just need to create a fn that calls what you already have, and returns an authentication map on success
19:40	cbp	howdy
19:40	headshot	howdy
19:40	cbp	anyone here preparing or planning to prepare for google codejam?
19:41	dxeh	cbp, google codejam?
19:41	ravster	I'm trying that, the code I have is at https://www.refheap.com/paste/11034 , but I'm getting a 'wrong number of args passed [0]' error.
19:41	headshot	coding competition
19:42	hyPiRion	cbp: Yeah
19:42	cbp	dxeh: http://code.google.com/codejam/
19:42	abp	shaunxcode: Sounds reasonable. Getting a satisfying cljs-environment in terms of libs etc. is a lot of work.
19:43	dxeh	cbp, reverse engineering?
19:43	DerGuteMoritz	http://dev.clojure.org/jira/browse/CLJS-467 <- if anyone is interested in my findings with regards to the bug I mentioned earlier
19:43	hyPiRion	dxeh: algorithm and data structures
19:44	cbp	I just solved a problem that was taking me over 2 hours with (memoize fn) and got excited
19:44	cbp	I was considering using a more iterative language for the contest but now I think i'll stick to clojure
19:44	ravster	cemerick: this function should return a map with {:name, :identity, :roles, etc}. That counts as a valid authentication map, right?
19:45	xeqi	ravster: are you useing `:credentials-fn friend-authenticate` ?
19:45	dxeh	im just picking up clojure, heard about it a year or 2 ago and have seen code snippets but never really looked into it
19:45	hyPiRion	cbp: I'd recommend going polyglot. I go Java for things like Floyd-Warshall and Dijkstra, but Clojure if there's combinatorics or permutations
19:45	dxeh	seems quite cool though
19:45	shaunxcode	abp: yeah, I am definitely heavily devoted to building the cljs ecosystem but I also want to "ship" features as I do that.
19:46	hyPiRion	Though I've realized you can get very far with just Clojure.
19:46	dnolen	DerGuteMoritz: thanks looks right to me, I've assigned the Michal to confirm
19:46	dnolen	assigned to
19:46	DerGuteMoritz	dnolen: alright, thank you!
19:46	ravster	xeqi: arrghh, no. I had that inside parens.
19:46	dnolen	DerGuteMoritz: please do send in your CA ;)
19:46	dnolen	DerGuteMoritz: that's a good catch, could use more of those.
19:46	ravster	xeqi: thanks :$ thats embarrassing
19:47	xeqi	ravster: np, happens to everyone at some point
19:47	cbp	hyPiRion: yes I'll probably do that
19:52	DerGuteMoritz	dnolen: yeah I'll try to do that as soon as possible, might take a moment though coming from Germany. on a different note, do you happen to know what license restrictions apply when I port code from cljs core to another language?
20:14	abp	Now there's an opportunity to beat the shit out of everyone who took the read-issue lightly (like me): (alter-var-root #'read-eval* (constantly :unknown)) on the top-level of every public ns of your libraries. :D
20:15	technomancy	abp: `lein new app mythingy` already includes that =)
20:15	amalloy	technomancy: i think you missed the joke there
20:16	abp	technomancy: read closely, and have a look at richs latest commits. :P
20:16	technomancy	uh oh
20:16	abp	Hm, I can't resist joking in here.
20:17	amalloy	abp: i'm not clear on how that's a more interesting "attack" than putting (/ 1 0) at the top level of all your namespaces, though
20:17	xeqi	I prefer ziping everything in src/ and sending it to a web server
20:17	abp	amalloy: Not attack, more in spirit of teaching a lesson. Secure your reads, a*
20:18	abp	amalloy: Via explicit read-eval-binding like loudly proposed
20:31	abp	Any pointers on how to start learning git, preferably with magit, if that makes any sense?
20:32	hiredman	abp: http://mirror.linux.org.au/pub/linux.conf.au/2013/ogv/Git_For_Ages_4_And_Up.ogv
20:32	technomancy	abp: magit has a screencast that's good
20:34	abp	hiredman: technomancy Thanks!
20:59	charlietanksley	abp: I don't know about learning git, but the magit user manual is pretty thorough: http://philjackson.github.com/magit/magit.html
21:00	charlietanksley	abp: I try to pick some random thing I find myself doing on via the cli every week or so, looking it up there, and trying to add just that to my magit workflow.
21:05	lynaghk	charlietanksley, abp: whenever I get into a new mode in emacs I always "C-h m" to see what commands are available. (I do pretty much everything in magit now.)
21:07	abp	charlietanksley: Sounds good, since I'm new to Emacs I try to learn it as broad of an replacement as possible. I haven't done much with git on the console anyway, so going magit from the beginning will work or not.
21:08	charlietanksley	lynaghk: good call.
21:09	charlietanksley	abp: I'm not sure how that will work. Good luck with it!
21:09	abp	lynaghk: sam is crazy, he's rebound C-h to delete in emacs live.
21:10	technomancy	eep
21:11	abp	Oh and made it M-h
21:13	abp	lynaghk: Thanks, thats really cool!
21:13	lynaghk	abp: yeah, no problem. it's the only emacs trick I really know, but you can use it to find out all of the other tricks =)
21:13	abp	Still can't believe I've procrastinated on learning Emacs until now...
21:14	cemerick	lynaghk: welcome back :-)
21:14	lynaghk	abp: you have taken the first step on the journey of a lifetime...
21:14	abp	Actually, people I've pushed into Clojure pushed me into Emacs in return.
21:14	lynaghk	cemerick: thanks! I'm so happy to be back in PDX. The Internet is now on a screen larger than 10 inches
21:14	technomancy	returning the favour =)
21:15	cemerick	lynaghk: were you gone long enough to start going a bit bonkers?
21:19	lynaghk	cemerick: yeah, especially in the more rural parts of New Zealand
21:20	lynaghk	cemerick: also traveling with the same person for six weeks, that was a bit intense. I did swing by SF on the way back for a week though and got to visit a lot of folks there
21:20	brehaut	lynaghk: whats 'more rural'?
21:21	lynaghk	cemerick: in San Francisco the Internet is, like, a real thing. With billboards and everything
21:21	lynaghk	brehaut: Riwaka on the South Island, which is a suburb of Moteuka
21:21	brehaut	lynaghk: i am from the south island ;)
21:21	lynaghk	brehaut: sorry I didn't get in touch, by the way. We were pretty busy for the days when we were in Auckland
21:21	brehaut	lynaghk: thats no problem at all
21:21	cemerick	brehaut: dude, don't trip over the village's internet wire
21:22	lynaghk	brehaut: ah, cool. Are you coming out to Clojure/west by any chance? A lot of the stuff I saw in the South Island reminded me a lot of Oregon.
21:22	brehaut	cemerick: theres only one. we are all very careful with it
21:22	brehaut	lynaghk: sadly no. way too far to travel
21:22	brehaut	i might make it to YOW!/Lambda Jam
21:22	cemerick	brehaut: my mental image is too hilarious for words
21:23	brehaut	cemerick: until kim dotcom builds a second cable, there is literally only one internet cable
21:23	brehaut	part of the reason everythign down here is slow as molasses
21:23	cemerick	really? I thought there was a whole aws region in NZ?
21:24	lynaghk	brehaut: I was talking with some nerds in Wellington and they said there was plenty of capacity, but that the provider basically just charged a ton for everything---whenever there is a credible threat that someone will build another cable the Internet cost mysteriously goes down
21:25	brehaut	lynaghk: yeah providers charging insane amounts is basically par for the course here
21:25	brehaut	not enough demand to create real competition
21:25	cemerick	oh, the AP region is in Sydney, my bad
21:26	tommo	what does # denote for example infront of a set?
21:26	brehaut	lynaghk: im not surprised riwka / moteka reminded you of oregan
21:26	brehaut	thats hops country
21:27	brehaut	(and riwaka is also a variety of hop)
21:28	brehaut	tommo: in front of a set? ##{:a :b}?
21:28	lazybot	⇒ {:a :b}
21:28	lynaghk	brehaut: re: beer, I'm tempted to buy a case of Moa's St. Joseph. That stuff was awesome
21:28	tommo	brehaut: yeah
21:28	cemerick	tommo: #{} is a set
21:28	cemerick	{} is a map
21:28	brehaut	lynaghk: its pretty good eh, i quite like their five hops in mid summer too
21:29	tommo	cemerick: ohh ok, i thought i seen them used elsewhere, that makes sense, ty
21:31	brehaut	lynaghk: did you get a chance to try any renaissance or 8wired beers there?
21:39	gfredericks	TimMc: hyPiRion: if neither of you are going to be at clojure/west I will submit a lightning talk on swearjure.
21:43	craigbro	new AUtechre album!
21:46	gfredericks	lein will not let me create a project in an existing directory? :/
21:55	TimMc	gfredericks: You have my blessing.
21:55	abp	O R G
21:57	abp	When I memoized those org-todo key-bindings ST2 travels into trash and everything's E
21:57	technomancy	=D
21:58	abp	brb, sleepersisting those bindings.
23:44	cemerick	bizarre; I think a particular rev of firefox is what was causing my ff/browser-repl problems a few weeks ago