#clojure – 2009-01-31

2009-01-31

00:06	eee	hi
00:06	eee	has anyone bought the 41 dollar book yet?
00:07	ayrnieu	I only paid for the PDF.
00:07	eee	that's all there is
00:07	eee	did it help you?
00:07	eee	i remember you . . did you already know lisp?
00:08	ayrnieu	thus far. IIRC, 41 dollars is the price both for the PDF (now and until it's finished and published) and the paper book.
00:08	eee	did you know lisp?
00:09	ayrnieu	I already knew lisp. The book has been pretty helpful -- the exploration of Java through clojure is well-done.
00:09	eee	i'm not buying that clojure is accessable to people that don't know lisp
00:10	hiredman	shrug
00:10	eee	I'm thinking of diggin into some lisp books first. good idea?
00:10	hiredman	clojure is a lisp, so knowing a lisp is required to know clojure...
00:10	eee	there's just too much people talk about that I don't grok
00:10	hiredman	clojurebot: sicp?
00:10	clojurebot	sicp is http://web.mit.edu/alexmv/6.001/sicp.pdf
00:11	hiredman	always good
00:11	hiredman	the hokey videos are nice too
00:11	ayrnieu	The Little Schemer will help you.
00:11	ozy`	eee: http://www.gigamonkeys.com/book/ <= if you're into pain
00:11	ozy`	(PCL is actually pretty gentle as far as I know)
00:12	ozy`	(scheme is better for learning though)
00:12	ozy`	(so, SICP)
00:12	eee	lol people always recommend those
00:12	eee	i guess they are standards at some schools
00:12	ozy`	eee: asked this question before, have you?
00:12	ozy`	PCL isn't used in schools, to my knowledge
00:12	eee	no
00:12	eee	not quite
00:13	eee	but people still recommend those
00:13	hiredman	the sicp videos are very gentle
00:13	eee	this question is whether you need to hit books like that before bothering with clojure . . . Rich has videos saying it's ok if clojure is your first sojourn into lisp . . . but I'm not finding that so far
00:14	Raynes	Why not recommend a book that is actually /about/ clojure. Like Programming Clojure. ;)
00:14	eee	there are videos?
00:14	hiredman	yeah
00:14	eee	well, I asked about that book
00:15	eee	raynes, it's pretty good?
00:15	ozy`	Raynes: he doesn't want to buy the book yet because he's afraid he'll have trouble learning clojure if he doesn't already know lisp
00:15	hiredman	eee: I always recomend hands on learning, the sooner you start writing code the better
00:15	ayrnieu	also, because he thinks it's 41 dollars.
00:15	eee	i do think it's 41 dollars :)
00:15	hiredman	,(println "hello world")
00:15	clojurebot	hello world
00:15	eee	yup
00:15	Raynes	eee: I wouldn't really know, I only read 20 pages of it because once I had to actually write a function I went back into my haskell-shell.
00:15	Raynes	And hid.
00:16	hiredman	man, talk about a low barrier to entry
00:16	eee	i can do plenty of simple stuff
00:16	ayrnieu	eee, the PDF is $21
00:16	eee	give me a simple challenge right now :) let's go
00:16	ozy`	,(map println '[clojure is neat])
00:16	clojurebot	(nil nil nil)
00:16	ozy`	aw man
00:17	eee	it is? see I have reading comprehension problem aymieu
00:17	ayrnieu	it's probably the horrible font you're using :-)
00:17	Raynes	Get emacs working seamlessly with Clojure on windows in 2 minutes. There is a challenge :p
00:17	eee	i did that
00:17	eee	someone has an easy setup for that
00:17	mfredrickson	,(doseq [i '[clojure is neat]] (print i))
00:18	clojurebot	clojureisneat
00:18	Raynes	Clojure-box doesn't count.
00:18	eee	i clicked a button and it installed . . .2 minutes exactly
00:18	eee	oh
00:18	mfredrickson	doseq vs map. i never get it right
00:18	hiredman	cb doesn't deal with line breaks well
00:18	hiredman	map is lazy
00:18	mfredrickson	either
00:18	hiredman	clojurebot: map?
00:18	clojurebot	map is LAZY
00:18	eee	but, emacs doesn't follow windows conventions of things like cut and paste and save
00:18	eee	who wants to relearn all that boilerplate . . . not to mention a new language
00:18	hiredman	and by the time cb forces the seq, it has alread closed out
00:18	Raynes	I use emacs for everything /but/ JVM languages, I can't handle it.
00:19	mfredrickson	is there a good windows vim? MacVim has been great to me
00:19	hiredman	,(doall (map pr (range 10)))
00:19	clojurebot	(nil nil nil nil nil nil nil nil nil nil)
00:19	clojurebot	0123456789
00:19	eee	oh and another thing . . . .this "advantage" of being able to use java stuff? well, I have programmed in java a bunch but never learned it. my IDE knows java, not me
00:20	eee	i never got macvim to work
00:20	eee	windows vim much easier
00:20	hiredman	,(.codePointAt "abcd" 0)
00:20	clojurebot	97
00:20	ozy`	eee: getting macvim to work is pretty easy... you just decompress it and open it
00:20	eee	i don't have a coding environment for clojure . . .BIG part of the problem
00:20	hiredman	that was a java call
00:20	ayrnieu	eee, I've loathed and avoided java for years, and Programming Clojure rehabilitated it for me.
00:21	hiredman	,(char 97)
00:21	Raynes	eee: The netbeans plugin is okay.
00:21	clojurebot	\a
00:21	eee	what did you use before aymieu?
00:21	Raynes	I stress that when I say "okay".
00:21	ayrnieu	,(map #(symbol (.getName %)) (.getMethods String))
00:21	clojurebot	(hashCode compareTo compareTo indexOf indexOf indexOf indexOf equals toString length isEmpty charAt codePointAt codePointBefore codePointCount offsetByCodePoints getChars getBytes getBytes getBytes getBytes contentEquals contentEquals equalsIgnoreCase compareToIgnoreCase regionMatches regionMatches startsWith startsWith endsWith lastIndexOf lastIndexOf lastIndexOf lastIndexOf substring substring subSequence concat replace
00:21	ayrnieu	eee - everything else.
00:21	ozy`	ayrnieu: I've thought for a long while that java would suck a lot less if only there was a better language to serve as the interface to its mediocre libraries
00:22	eee	i can make a clojure program as long as there are no dependencies and in a single file in the netbeans plugin
00:22	hiredman	for clojurebot developement
00:22	lisppaste8	ayrnieu pasted "Java-using MD5 " at http://paste.lisp.org/display/74629
00:22	hiredman	clojurebot: emacs?
00:22	clojurebot	"Learning Emacs and SLIME was just about the most painful computer experience I've ever been through."
00:22	Raynes	If VIM were the last editor of any type on the planet. I wouldn't code.
00:22	ozy`	hiredman: ah, so you're the one....
00:23	eee	i agree with the bot
00:23	ayrnieu	I didn't try to be clever with that; it's just the first example I saw online of someone writing an MD5 in Java.
00:23	Raynes	The whole classpath bullshit really lowered my clojure-excitement.
00:24	eee	i dropped an AI class I really wanted to take because I wasted 48 hours straight trying to get emacs and a repl going for lisp
00:24	eee	raynes I agree
00:24	hiredman	clojurebot: jar directory?
00:24	eee	I hate all that classpath crap
00:24	clojurebot	jar directory is find .jars/ -type f -name \*.jar -print0\|xargs -0\|sed "s/ /:/g"
00:24	ozy`	Raynes: you know you can set up emacs keybindings in vim, right?
00:24	eee	ozy, then why use vim?
00:25	Raynes	ozy`: I don't like vim, I don't want to make it more like emacs, I don't want to make it /act/ like emacs.
00:25	hiredman	ozy`: emacs goes a bit beyond keybindings
00:25	Raynes	I want emacs.
00:25	Raynes	:\|
00:25	ozy`	eee: oh, I was just being pedantic with his rhetorical situation :p
00:25	eee	i want type ahead and dot completion
00:25	ayrnieu	Raynes, the whole classpath bullshit is familiar to me from dealing with C libraries, ldconfig, etc. You just build a little bit of architecture to handle it, and continue. I 'exec zsh -l' to refresh my CLASSPATH -- I could be cleverer.
00:25	ozy`	eee: you have no soul
00:25	Raynes	tl;dr ;)
00:26	hiredman	eee: I think there just was a message on the google group about a netbeans something or other with completion
00:26	eee	here's the FUNNIEST thing
00:26	eee	i saw that
00:26	eee	excited to try it
00:26	eee	this is funny
00:26	eee	tho
00:26	eee	ready?
00:26	eee	people talk about functional languages
00:26	eee	nice
00:26	hiredman	ayrnieu: zsh, Nice.
00:26	eee	high level
00:26	eee	right?
00:27	eee	and then they TOIL setting up and playing with environments and writing scropts
00:27	eee	the other thing
00:27	eee	iss
00:27	eee	is
00:27	ozy`	eee: and you're astonished that they're all so primitive that they lack basic stuff like for loops?
00:27	eee	eclipse writes all the boilerplate for you in java
00:27	eee	so it's not fair to say that lisp programs are shorter
00:27	eee	what percentage do I actually write?
00:27	eee	not much
00:27	hiredman	clojurebot: what do you know about trolls?
00:27	clojurebot	trolls are like harmless particles that people have terrible allergies to. If you don't start shouting and blowing your nose at them, they're just misguided young men.
00:27	eee	eclipse writes it
00:28	ayrnieu	eee, eclipse doesn't read all the boilerplate for you.
00:28	mfredrickson	eclipse writes it, you maintain it
00:28	ozy`	eee: on the other hand, what percentage do you actually read? answer: you have to read -all- of it.
00:28	mfredrickson	ozy`: +1
00:28	eee	well
00:28	eee	I can't read someone elses lisp thats for sure
00:28	ozy`	eee: always assume that your program will be written once, updated ten times, and read a hundred times
00:29	mfredrickson	to argue the other side of the coin, it's not eee's fault that java is verbose
00:29	eee	ever read output from a genetic program?
00:29	ozy`	eee: heh, well you don't know lisp, do you?
00:29	mfredrickson	best of a bad situation
00:29	eee	I think ou can write 10 lines of java in the time it takes to figure out your one line of lisp sometimes
00:30	eee	and eclipse let's you collapse the little plusses :)
00:30	eee	but I really appreciate you helping me with these arguments
00:30	ozy`	well that's definitely a plus ;)
00:30	eee	because I have to convince coworkers to try functional languages more
00:30	eee	so I'm being them :)
00:31	eee	here's a real question
00:31	eee	someone wrote in the discussions
00:31	eee	that they made set ops faster
00:31	ozy`	learning a functional language will help you understand java's whacked-out design patterns better... a lot of them are just emulating stuff you get for free in functional langs
00:31	ayrnieu	it's better to seek answers with questions than to seek resistance with opinions.
00:32	ozy`	ayrnieu: that should go on a plaque over the entrance to 4chan or something
00:32	eee	well, would it be faster to drop into empiricle mode? copy everything to a java set, then call java's intersect stuff, then copy back into a list?
00:32	hiredman	hmmm, picking unicode symbols for macros then trying to edit the file again in a non-unicode console
00:32	hiredman	not the best plan I have ever made
00:33	ayrnieu	eee, clojure doesn't just use lists - actually it's remarkable for how well it uses other data types.
00:33	hiredman	,({:a 1 :b 2} :a)
00:33	clojurebot	1
00:34	ayrnieu	eee, c.f. Chapter 4: Unifying Data with Sequences, of the $21 book :-)
00:34	eee	i see that now
00:35	eee	does the book talk about web stuff, too?
00:35	ayrnieu	so far it doesn't.
00:36	ayrnieu	There's a 'web development' section in Chapter 9.
00:36	eee	i got one of those slicehosts
00:36	eee	so I could follow eric lavigne's tutprial
00:36	eee	I just did it by wrote
00:36	eee	and of course it worked
00:36	eee	but I didn't grok it :)
00:37	eee	I can do simple little algorithms
00:37	eee	that's it
00:37	eee	:)
00:37	eee	anyone got a little challenge?
00:38	hiredman	eee: you could hit up project euler
00:38	hiredman	they have a few challenges
00:38	briancarper	"Learning Emacs and SLIME... most painful experience..." did I just hear myself quoted by clojurebot or did someone else say that too?
00:38	eee	lol
00:39	hiredman	briancarper: could have been you
00:39	hiredman	that was your blog post?
00:39	briancarper	Yeah.
00:39	hiredman	Yeah.
00:39	eee	where are these movies someone mentioned?
00:39	ayrnieu	'screencasts' on clojure.org
00:39	hiredman	clojurebot: google sicp videos
00:39	mfredrickson	or did you mean SICP
00:39	clojurebot	First, out of 17600 results is:
00:40	clojurebot	Structure and Interpretation of Computer Programs, Video Lectures
00:40	clojurebot	http://groups.csail.mit.edu/mac/classes/6.001/abelson-sussman-lectures/
00:40	briancarper	Those movies are good.
00:40	hiredman	clojurebot: ?
00:40	clojurebot	It's greek to me.
00:40	eee	i saw rich's clojure ones. cool . . .the lectures
00:40	eee	thanks
00:41	briancarper	I like the parts where Sussman types LISP laboriously on an old green-on-black 30-character-wide display.
00:41	hiredman	I put them on my blackberry
00:41	hiredman	briancarper: ha ha
00:41	hiredman	the magic parens balancing
00:41	hiredman	the hats
00:41	briancarper	His huge sidewalk chalk.
00:42	eee	did y'all go to MIT or wherever that was?
00:42	hiredman	eee: no, I watched the videos
00:43	hiredman	berkely had a podcast of the audio from the first three of their CS courses
00:43	eee	that's cool
00:43	hiredman	the first is scheme, the second is java, the third is C and asm
00:44	hiredman	those are just he languages the classes use, the classes are not about those languages
00:44	hiredman	the C/asm class was awesome
00:45	eee	C/asm was your third course?
00:46	eee	oh I see
00:46	hiredman	clojurebot: google berkely 61c mp3s
00:46	clojurebot	First, out of 1770 results is:
00:46	clojurebot	CS 61C Machine Structures Introduction
00:46	clojurebot	http://www.jimmyr.com/blog/CS_61C_Machine_Structures_Lecture_1_John_Wawrzynek_159_2006.php
00:46	eee	that was an aside
00:46	hiredman	eee: I did not personally attend berkely
00:47	eee	i see
00:47	hiredman	I have taken CS classes at the local community college, but they are a joke
00:48	eee	you taught yourself?
00:48	hiredman	it just made me really bitter
00:48	hiredman	eee: I never think of that way
00:49	hiredman	I just know how to read instructions
00:49	hiredman	and watch videos
00:49	eee	i'm the opposite
00:49	eee	i need to start out from someone else
00:49	hiredman	many people often seem to be the opposite
00:49	eee	i like to use people like compilers
00:49	dreish	What are the primitives, and what are the means of combinaaation?
00:50	hiredman	eee: project euler has all kinds of problems to take a crack at
00:50	eee	i liked this book: http://www.amazon.com/Structures-Algorithms-CBMS-NSF-Conference-Mathematics/dp/0898711878
00:50	hiredman	it was fun while I was doing it, maybe I should start again
00:50	hiredman	dreish: word
00:50	eee	i'll check out project euler
00:52	hiredman	eee: the first problem is, find the sum of all the multiples of 3 or 5 under 1000
00:52	eee	hmmmmmm
00:52	eee	ok
00:53	eee	see, interestingly . . . my mind starts thinking, "while"
00:54	eee	which isn't really used in clojure
00:55	eee	so I guess a silly way to solve it might be to just loop from 1 to 1000
00:55	eee	and
00:55	eee	if a number evenly divides by 3 or 5
00:55	eee	add it to some list
00:55	eee	then reduce the list
00:55	hiredman	a lot of the time things that would be some kind of loop in another language are seqs in clojure
00:56	dreish	Also, there is no list.
00:56	eee	so evenly divides is checked by having mod 3 equal to zero
00:56	dreish	When you create a seq, there's usually just one or two things in memory at a time.
00:56	ozy`	eee: think less about what steps to take to get the answer, and more about how to describe the answer in terms of its inputs
00:57	dreish	,(apply + (range 1000000))
00:57	clojurebot	499999500000
00:57	eee	ahhhhhh!
00:57	eee	i got beat
00:57	eee	ok
00:57	eee	how do that as reduce?
00:57	dreish	,(reduce + (range 1000000))
00:57	hiredman	don't use reduce
00:57	clojurebot	499999500000
00:58	eee	why not? i've never used it
00:58	hiredman	ok, go ahead
00:58	eee	:)
00:58	cp2	lol
00:58	ayrnieu	,(interpose + (range 10))
00:58	clojurebot	(0 #<core$_PLUS___3183 clojure.core$_PLUS___3183@1b0d990> 1 #<core$_PLUS___3183 clojure.core$_PLUS___3183@1b0d990> 2 #<core$_PLUS___3183 clojure.core$_PLUS___3183@1b0d990> 3 #<core$_PLUS___3183 clojure.core$_PLUS___3183@1b0d990> 4 #<core$_PLUS___3183 clojure.core$_PLUS___3183@1b0d990> 5 #<core$_PLUS___3183 clojure.core$_PLUS___3183@1b0d990> 6 #<core$_PLUS___3183 clojure.core$_PLUS___3183@1b0d990> 7 #<core$_PLUS___3183 clo
00:58	cp2	hiredman, i am curious though
00:58	cp2	why not?
00:58	hiredman	eee: make an account on the euler website
00:58	ayrnieu	,(interpose '+ (range 10)) ;; readable mode
00:58	clojurebot	(0 + 1 + 2 + 3 + 4 + 5 + 6 + 7 + 8 + 9)
00:58	eee	how do you use reduce now?
00:58	ozy`	cp2: because '+ is already a fold, I assume....
00:59	cp2	oh well, i wasnt paying attention to the topic
00:59	eee	how do you check docs on something?
00:59	dreish	Topic is Euler puzzle 1.
00:59	cp2	my bad, just say hiredman saying not to use something, and it sparked interest. didnt care to read further ;)
00:59	ayrnieu	,(doc interleave)
00:59	clojurebot	"([& colls]); Returns a lazy seq of the first item in each coll, then the second etc."
00:59	eee	oh, reduce was the same as apply
01:00	eee	that's no fun
01:00	hiredman	no
01:00	ayrnieu	the same use and the same effect, but they differ.
01:00	hiredman	reduce and apply are different
01:00	ozy`	eee: apply only has the same effect if you use it on something that already implies 'reduce
01:00	eee	and then there's "map"
01:00	dreish	,(apply map #(+ 2) '((1 2)))
01:00	clojurebot	java.lang.IllegalArgumentException: Wrong number of args passed to: sandbox$eval--1372$fn
01:01	dreish	ur
01:01	ozy`	eee: 'map is easier to grok. you'll get back a sequence exactly as long as the one you passed into it
01:01	dreish	,(apply map #(+ % 2) '((1 2)))
01:01	clojurebot	(3 4)
01:01	dreish	,(reduce map #(+ % 2) '((1 2)))
01:01	clojurebot	(3 4)
01:01	eee	,(reduce (map '+ (range 5)))
01:01	clojurebot	java.lang.IllegalArgumentException: Wrong number of args passed to: core$reduce
01:01	hiredman	eee: sign up for an account on the euler website, they have a list of problems
01:01	hiredman	some of them are rather difficult
01:02	hiredman	spend sometime with clojure.org/api
01:02	ayrnieu	,(let [calls (ref 0) my+ (fn [a b] (dosync (alter calls inc)) (+ a b))] [(apply my+ (range 10000)) @calls])
01:02	clojurebot	java.lang.IllegalArgumentException: Wrong number of args passed to: sandbox$eval--1395$my-PLUS-
01:02	eee	dude clojure api is so terse. it's written for people who use very few . .. but big words
01:02	eee	like "intern"
01:03	ayrnieu	oh, duh.
01:03	eee	and "destructure"
01:03	eee	i'm like, "thanks. that helps"
01:03	cp2	eee its better than garbled abrreviations/acronyms
01:03	ozy`	,(apply + (map #(apply + (filter (fn [k] (= (mod % k) 0)) (range 10000))) '(3 5)))
01:03	clojurebot	java.lang.RuntimeException: java.lang.RuntimeException: java.lang.ArithmeticException: Divide by zero
01:03	ayrnieu	easy to fix, but nevermind.
01:03	ozy`	whoops
01:03	ayrnieu	eee, ask questions in the channel.
01:03	hiredman	eee: the sequence functions are very useful
01:04	cp2	ozy`: im glad you dont work on the lhc
01:04	ozy`	,(apply + (map #(apply + (filter (fn [k] (= (mod k %) 0)) (range 10000))) '(3 5)))
01:04	clojurebot	26663333
01:04	hiredman	eee: anyway, you asked for a challenge, so have at euler
01:04	eee	that's pretty cool, no if statements were used
01:04	eee	yes, thankyou hiredman
01:05	ozy`	eee: 'filter contains an if statement, IIRC
01:05	LordOfTheNoobs	clojurebot: Peter Griffin?
01:05	clojurebot	You and your big words and your small difficult words
01:05	eee	i bet filter could be avoided by looking up the list comprehension syntax
01:05	ayrnieu	eee, find me a hundred pairs of numbers whose base-seven product contains the number of the beast
01:05	ozy`	eee: anyway I think my implementation is wrong given the problem specification
01:06	eee	i know filter does . . .you guys helped me make "separate" faster than what's in contrib last time
01:06	hiredman	ozy`: that certainly is not the right answer
01:06	ozy`	cp2: anyway I didn't think it would even parse... wasn't too careful about balancing parentheses
01:06	dreish	(+ (apply + (range 0 1000 3)) (apply + (range 0 1000 5)) (apply - (range 0 1000 15)))
01:06	dreish	That's almost like cheating.
01:06	ozy`	hiredman: not too worried :p just playing around
01:06	hiredman	clojurebot: clojure?
01:06	clojurebot	clojure is a very attractive hammer with a nice heft to it
01:07	hiredman	clojurebot: clojure?
01:07	clojurebot	clojure is a very attractive hammer with a nice heft to it
01:07	hiredman	grr
01:07	ozy`	dreish: ahhh didn't think of that one
01:07	hiredman	clojurebot: clojure?
01:07	clojurebot	clojure is cheating
01:07	hiredman	that's the one
01:07	ayrnieu	clojurebot, use non-repeating RNG
01:08	briancarper	Is there any way to break an infinite loop at the REPL without killing the whole Clojure process?
01:08	hiredman	ayrnieu: sometimes you get a bunch of heads in a row when you flip a coin
01:08	eee	wow. that's smart dreish
01:08	ozy`	> foldl' (+)
01:08	ozy`	aaagghhh
01:08	hiredman	no lambdabot here
01:09	ozy`	<_<
01:09	ozy`	>_>
01:10	eee	i'll have to think about yours for a long time aymieu
01:10	briancarper	I'll take lack of response to my question as a no.
01:11	eee	i have had the same problem in enclojure
01:11	eee	and had to kill netbeans
01:11	eee	let alone the repl
01:12	dreish	briancarper: I haven't found one, but I think it should be possible to do.
01:12	briancarper	You'd think.
01:14	hiredman	try kill -QUIT the java process
01:14	eee	(for seq-exprs expr) Macro List comprehension. Takes a vector of one or more binding-form/collection-expr pairs, each followed by an optional filtering :when/:while expression (:when test or :while test), and yields a lazy sequence of evaluations of expr. Collections are iterated in a nested fashion, rightmost fastest, and nested coll-exprs can refer to bindings created in prior binding-forms. (take 100 (for [x (range 100000
01:14	eee	see
01:14	ayrnieu	briancarper, how much do you want to preserve from the repl?
01:14	eee	it doesn't say what when does
01:14	hiredman	that might just make the threads dump there stack
01:14	eee	verses while
01:14	eee	just says you can use one or the other
01:14	hiredman	eee: while is while exp is true
01:14	hiredman	when is when expr is true
01:15	eee	um . .. in a list comprehension that makes no sense
01:15	hiredman	it totally does
01:15	eee	haskel list comprehension makes perfect sense
01:15	eee	python list comprehension makes totl sense
01:15	eee	you state the predicate
01:15	eee	this when thing is meaningless
01:15	ayrnieu	,(for [x (range 100) :when (zero? (mod x 3))] x)
01:15	clojurebot	(0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 66 69 72 75 78 81 84 87 90 93 96 99)
01:16	briancarper	ayrnieu: I want to kill the loop and go back to the repl without losing everything in the environment.
01:16	ayrnieu	if you want a different syntax, write it.
01:16	eee	,(for [x (range 100) :while (zero? (mod x 3))] x)
01:16	clojurebot	(0)
01:16	eee	ok
01:16	eee	totally arbitrary
01:16	eee	but ok
01:16	hiredman	,(for [x (range 1000) :when (and (zero? (mod x 3)) (zero? (mode x 5)))] x)
01:16	clojurebot	java.lang.Exception: Unable to resolve symbol: mode in this context
01:17	ayrnieu	what's totally arbitrary?
01:17	LordOfTheNoobs	eee: while stops at the first failure when moving over a list, when samples through the list. I think ( checking )
01:17	hiredman	,(for [x (range 1000) :when (or (zero? (mod x 3)) (zero? (mod x 5)))] x)
01:17	clojurebot	(0 3 5 6 9 10 12 15 18 20 21 24 25 27 30 33 35 36 39 40 42 45 48 50 51 54 55 57 60 63 65 66 69 70 72 75 78 80 81 84 85 87 90 93 95 96 99 100 102 105 108 110 111 114 115 117 120 123 125 126 129 130 132 135 138 140 141 144 145 147 150 153 155 156 159 160 162 165 168 170 171 174 175 177 180 183 185 186 189 190 192 195 198 200 201 204 205 207 210 213 215 216 219 220 222 225 228 230 231 234 235 237 240 243 245 246 249 250 252
01:17	ayrnieu	,(for [x (range 30) :when (zero? (mod x 3)) y (range 30) :while (< (* 2 x) 30)] [x y])
01:17	clojurebot	([0 0] [0 1] [0 2] [0 3] [0 4] [0 5] [0 6] [0 7] [0 8] [0 9] [0 10] [0 11] [0 12] [0 13] [0 14] [0 15] [0 16] [0 17] [0 18] [0 19] [0 20] [0 21] [0 22] [0 23] [0 24] [0 25] [0 26] [0 27] [0 28] [0 29] [3 0] [3 1] [3 2] [3 3] [3 4] [3 5] [3 6] [3 7] [3 8] [3 9] [3 10] [3 11] [3 12] [3 13] [3 14] [3 15] [3 16] [3 17] [3 18] [3 19] [3 20] [3 21] [3 22] [3 23] [3 24] [3 25] [3 26] [3 27] [3 28] [3 29] [6 0] [6 1] [6 2] [6 3]
01:17	eee	,(for [x (range 100) :while (not (zero? (mod x 3)))] x)
01:17	clojurebot	nil
01:18	dreish	You really don't want to type :when, do you?
01:18	LordOfTheNoobs	,(mod 0 3)
01:18	clojurebot	0
01:18	eee	i always want to type when
01:18	ayrnieu	,(for [x (range 30) :when (zero? (mod x 3)) y (range 30) :while (< (* 2 y) 30)] [x y])
01:18	clojurebot	([0 0] [0 1] [0 2] [0 3] [0 4] [0 5] [0 6] [0 7] [0 8] [0 9] [0 10] [0 11] [0 12] [0 13] [0 14] [3 0] [3 1] [3 2] [3 3] [3 4] [3 5] [3 6] [3 7] [3 8] [3 9] [3 10] [3 11] [3 12] [3 13] [3 14] [6 0] [6 1] [6 2] [6 3] [6 4] [6 5] [6 6] [6 7] [6 8] [6 9] [6 10] [6 11] [6 12] [6 13] [6 14] [9 0] [9 1] [9 2] [9 3] [9 4] [9 5] [9 6] [9 7] [9 8] [9 9] [9 10] [9 11] [9 12] [9 13] [9 14] [12 0] [12 1] [12 2] [12 3] [12 4] [12 5] [1
01:20	ayrnieu	eee - here, implement max with list comprehensions.
01:21	eee	ok
01:21	eee	then I'd use while
01:21	eee	?
01:21	LordOfTheNoobs	, {:while (for [aa (range 1 10) :while (not (= aa 5))] aa) :when (for [aa (range 1 10) :when (not (= aa 5))] aa)}
01:21	clojurebot	{:while (1 2 3 4), :when (1 2 3 4 6 7 8 9)}
01:22	eee	what is that?
01:22	ayrnieu	what is what?
01:23	eee	what LON wrote
01:23	ayrnieu	,*1
01:23	clojurebot	java.lang.IllegalStateException: Var clojure.core/*1 is unbound.
01:23	eee	i'm looking up haskell right now
01:23	eee	seeing how they do it
01:24	ayrnieu	,[(class {:a 1 :b 2}), (class :a), (class (for [x (range 4)] x))]
01:24	clojurebot	[clojure.lang.PersistentArrayMap clojure.lang.Keyword clojure.lang.LazyCons]
01:24	LordOfTheNoobs	A hash-table literal with two similar for-expression calls to demonstrate their difference.
01:25	eee	doesn't seem like list comprehension would be efficient for max
01:25	eee	since you have to look at all the numbers
01:25	eee	i can only think of empricle implementatin
01:25	ayrnieu	no, it would not be efficient.
01:26	eee	i'm drawing a blank on that one
01:27	hiredman	ayrnieu: huh?
01:27	ayrnieu	here's the answer in set notation: { x \| x E l ( no y E l > x ) }
01:27	eee	sure
01:27	hiredman	oh, right
01:28	eee	need random
01:28	eee	,(rand-int 30)
01:28	clojurebot	5
01:28	ayrnieu	what needs random?
01:28	eee	cool
01:28	eee	i want to test on randome sequence
01:28	eee	,(take 30 (rand-int 30))
01:28	clojurebot	java.lang.IllegalArgumentException: Don't know how to create ISeq from: Integer
01:29	eee	,(take 30 (repeatedly (rand-int 30)))
01:29	clojurebot	java.lang.RuntimeException: java.lang.ClassCastException: java.lang.Integer cannot be cast to clojure.lang.IFn
01:29	eee	,(take 30 (repeatedly '(rand-int 30)))
01:29	clojurebot	java.lang.RuntimeException: java.lang.ClassCastException: clojure.lang.PersistentList cannot be cast to clojure.lang.IFn
01:29	ayrnieu	,(apply str (take 100 (repeatedly #(char (+ (int \a) (rand-int 26))))))
01:29	clojurebot	"zblkucvjrxtwgxeuxfzvxpnaokiimmvpugdxulqeketawsnzjdndeszrzwojpvfqjgjgdsgjvsuibjqnubbortiamajdcvehksta"
01:30	eee	,(take 30 (repeatedly #(rand-int 30)))
01:30	clojurebot	(15 20 10 26 0 14 7 9 25 12 9 1 3 28 23 9 6 8 5 18 0 9 5 17 26 25 4 14 25 28)
01:30	eee	ok
01:31	ayrnieu	,(.contains (apply str (take 1000000 (repeatedly #(char (+ (int \a) (rand-int 26)))))) "ayrnieu")
01:31	clojurebot	false
01:32	eee	,(for [x (take 30 (repeatedly #(rand-int 30))) :while
01:32	clojurebot	Eval-in-box threw an exception:EOF while reading
01:32	eee	i dunno
01:32	eee	need need two variable
01:33	eee	you have them in your math
01:33	eee	y and x
01:35	eee	,(for [x (take 30 (repeatedly #(rand-int 30))) y x] :while (< y x))
01:35	clojurebot	java.lang.IllegalArgumentException: Wrong number of args passed to: core$for
01:35	eee	,(for [x (take 30 (repeatedly #(rand-int 30))) y x :while (< y x) [x])
01:35	clojurebot	Eval-in-box threw an exception:Unmatched delimiter: )
01:36	eee	,(for [x (take 30 (repeatedly #(rand-int 30))) y x :while (< y x)] [x])
01:36	clojurebot	java.lang.IllegalArgumentException: Don't know how to create ISeq from: Integer
01:36	ayrnieu	why are you using while?
01:36	eee	,(for [x (take 30 (repeatedly #(rand-int 30))) y x :while (< y x)] x)
01:36	clojurebot	java.lang.IllegalArgumentException: Don't know how to create ISeq from: Integer
01:37	eee	trying to stop on the max
01:37	eee	i dunno . . .shouldn't be a while or a when
01:37	eee	it's set theory
01:37	eee	shold just be what's in the set or what's not
01:38	ayrnieu	well, here's how I put it: { x \| x E l ( no y E l > x ) }
01:38	hiredman	your syntax for for is incorrect
01:38	eee	i see another problem
01:38	eee	can't do it in this environment with a random sequence
01:38	eee	oh
01:38	ayrnieu	( no y E l > x ) stipulates that no element of l is greater than x.
01:38	eee	unless I use let
01:38	eee	i need to do a let to hold the randome sequence so I can use it with x and y
01:39	eee	that way I can have what will amount to a nested loop
01:39	eee	or maybe nested loop happens too
01:39	eee	oh yeah
01:39	eee	all combinations of x and y can be tried
01:40	eee	somehow
01:40	hiredman	,(doc for)
01:40	clojurebot	"([seq-exprs expr]); List comprehension. Takes a vector of one or more binding-form/collection-expr pairs, each followed by an optional filtering :when/:while expression (:when test or :while test), and yields a lazy sequence of evaluations of expr. Collections are iterated in a nested fashion, rightmost fastest, and nested coll-exprs can refer to bindings created in prior binding-forms. (take 100 (for [x (range 100000000
01:40	hiredman	^- "collections are iterated in a nested fashion"
01:41	hiredman	hah
01:42	eee	,(let [myseq (take 30 (repeatedly #(rand-int 30)))] myseq)
01:42	clojurebot	(4 17 16 8 16 23 25 16 28 23 4 7 2 17 16 10 13 21 25 2 7 24 15 7 3 22 3 1 20 1)
01:43	eee	,(let [myseq (take 30 (repeatedly #(rand-int 30)))] (for [x myseq] x)
01:43	clojurebot	Eval-in-box threw an exception:EOF while reading
01:43	eee	,(let [myseq (take 30 (repeatedly #(rand-int 30)))] (for [x myseq] x))
01:43	clojurebot	(11 17 4 19 9 24 8 6 1 16 6 0 11 10 9 5 25 27 7 0 5 1 7 6 25 2 14 18 0 29)
01:43	eee	ok
01:43	eee	,(let [myseq (take 30 (repeatedly #(rand-int 30)))] (for [x myseq y myseq] (x y)))
01:43	clojurebot	java.lang.RuntimeException: java.lang.ClassCastException: java.lang.Integer cannot be cast to clojure.lang.IFn
01:43	eee	,(let [myseq (take 30 (repeatedly #(rand-int 30)))] (for [x myseq y myseq] [x y]))
01:43	clojurebot	([29 29] [29 2] [29 28] [29 11] [29 3] [29 11] [29 7] [29 8] [29 27] [29 18] [29 26] [29 15] [29 5] [29 13] [29 20] [29 18] [29 11] [29 9] [29 6] [29 29] [29 11] [29 18] [29 16] [29 8] [29 16] [29 18] [29 3] [29 21] [29 27] [29 25] [2 29] [2 2] [2 28] [2 11] [2 3] [2 11] [2 7] [2 8] [2 27] [2 18] [2 26] [2 15] [2 5] [2 13] [2 20] [2 18] [2 11] [2 9] [2 6] [2 29] [2 11] [2 18] [2 16] [2 8] [2 16] [2 18] [2 3] [2 21] [2 27]
01:44	eee	,(let [myseq (take 30 (repeatedly #(rand-int 30)))] (for [x myseq y myseq] x))
01:44	clojurebot	(5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 26 26 26 26 26 26 26 26 26 26 26 26 26 26 26 26 26 26 26 26 26 26 26 26 26 26 26 26 26 26 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 29 29 29 29 29 29 29 29 29 29 29 2
01:44	eee	,(let [myseq (take 30 (repeatedly #(rand-int 30)))] (for [x myseq y myseq] :while (> x y)))
01:44	clojurebot	java.lang.IllegalArgumentException: Wrong number of args passed to: core$for
01:44	eee	lost
01:44	eee	,(let [myseq (take 30 (repeatedly #(rand-int 30)))] (for [x myseq y myseq] :when (> x y)))
01:44	clojurebot	java.lang.IllegalArgumentException: Wrong number of args passed to: core$for
01:45	eee	,(let [myseq (take 30 (repeatedly #(rand-int 30)))] (for [x myseq y myseq :when (> x y)] x))
01:45	clojurebot	(13 13 13 13 13 13 13 13 13 13 13 12 12 12 12 12 12 12 12 12 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 28 28 28 28 28 28 28 28 28 28 28 28 28 28 28 28 28 28 28 28 28 28 28 28 28 28 28 28 8 8 8 8 8 8 12 12 12 12 12 12 12 12 12 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 20 20 20 20 20 20 20 20 20 20 20 20 20 20 2
01:45	eee	,(let [myseq (take 30 (repeatedly #(rand-int 30)))] (for [x myseq y myseq :while (> x y)] x))
01:45	clojurebot	(29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 28 28 28 28 28 28 28 28 28 28 28 28 28 28 28 28 28 28 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27)
01:45	durka42	what would be the fastest reference type for an integer that has to be modified thousands of times very frequently, but no synchronization is needed
01:45	eee	this must be painful to watch :)
01:46	hiredman	durka42: you can use commute instead of alter?
01:46	eee	you can use with-local-vars, maybe?
01:46	durka42	i'm not sure it's commutative
01:47	hiredman	oh
01:47	hiredman	I misread
01:47	durka42	my function gets mapped over a giant array and i need the max value
01:47	hiredman	,(doc atom)
01:47	clojurebot	"([x] [x & options]); Creates and returns an Atom with an initial value of x and zero or more options (in any order): :meta metadata-map :validator validate-fn If metadata-map is supplied, it will be come the metadata on the atom. validate-fn must be nil or a side-effect-free fn of one argument, which will be passed the intended new state on any state change. If the new state is unacceptable, the validate-fn should
01:47	durka42	that's what i'm using now
01:47	durka42	alternatively, a divide-and-conquer approach might be faster
01:47	durka42	if i can wrest control of the giant array away from quicktime
01:48	hiredman	hmmm
01:48	ayrnieu	,(time (let [x (ref 0)] (dorun (take 1000 (repeatedly #(dosync (alter x inc))))) @x))
01:48	clojurebot	1000
01:48	clojurebot	"Elapsed time: 49.594 msecs"
01:48	durka42	most approaches would be faster than calling a function (* 320 240) times, 16 times a second
01:48	eee	i don't trust the timer in clojure at all
01:48	durka42	,(time (let [x (atom 0)] (dorun (take 1000 (repeatedly (swap! x inc)))) @x))
01:48	clojurebot	java.lang.RuntimeException: java.lang.RuntimeException: java.lang.ClassCastException
01:48	durka42	why not?
01:49	ayrnieu	eee - dorun
01:49	eee	the function we wrote the other day was way faster
01:49	eee	but someone else had to fiigure out how to prove that
01:49	eee	nope
01:49	ayrnieu	,(time (let [x (ref 0)] (dorun (take 10000000 (repeatedly #(dosync (alter x inc))))) @x))
01:49	durka42	,(time (let [x (atom 0)] (dorun (take 1000 (repeatedly #(swap! x inc)))) @x))
01:49	eee	we actually had to break up the algorithms
01:49	clojurebot	Execution Timed Out
01:49	clojurebot	1000
01:49	clojurebot	"Elapsed time: 158.512 msecs"
01:49	eee	and dorun parts of it individually
01:49	durka42	ok, refs are faster
01:49	durka42	apparently
01:49	durka42	,(time (let [x (agent 0)] (dorun (take 1000 (repeatedly #(send x inc)))) @x))
01:49	clojurebot	0
01:49	clojurebot	"Elapsed time: 8.984 msecs"
01:49	eee	try with-local-vars
01:50	durka42	(doc with-local-vars)
01:50	clojurebot	varbinding=> symbol init-expr Executes the exprs in a context in which the symbols are bound to vars with per-thread bindings to the init-exprs. The symbols refer to the var objects themselves, and must be accessed with var-get and var-set; arglists ([name-vals-vec & body])
01:50	eee	you can just use @
01:50	ayrnieu	durka, @x isn't necessarily inc'd by the end
01:51	durka42	whoops
01:51	eee	so aymieu, do you know the answer to the challenge you gave me? was I close?
01:51	durka42	refs are not at all faster
01:52	ayrnieu	eee - use a second list comprehension to perform the 'no y greater than x' test.
01:52	ayrnieu	nested comprehensions. Horrid complexity.
01:52	eee	i was thinking that . . . .but was trying to avoid it
01:52	eee	forget it
01:52	eee	going to bed
01:53	eee	thanks for the schooling tho
01:53	durka42	,(time (let [x (agent 0)] (dorun (take 1000 (repeatedly #(send x inc)))) (await) @x))
01:53	clojurebot	0
01:53	clojurebot	"Elapsed time: 9.308 msecs"
01:53	durka42	,(time (let [x (agent 0)] (dorun (take 1000 (repeatedly #(send x inc)))) (await x) @x))
01:53	clojurebot	1000
01:53	clojurebot	"Elapsed time: 264.598 msecs"
01:53	durka42	aah
01:53	durka42	this is hopeless and un-clojure-like
01:54	durka42	i shall have to parallelize it
01:54	eee	what if you use java
01:55	eee	i mean for that part
01:55	durka42	i might have to do that
01:55	durka42	eh?
01:55	durka42	Gorilla=> (use 'clojure.parallel)
01:55	durka42	java.lang.ClassNotFoundException: jsr166y.forkjoin.ParallelArray (parallel.clj:32)
01:55	hiredman	durka42: you should just loop/recur
01:55	durka42	that's the problem! it's not my loop
01:55	durka42	quicktime gives me an interface to extend
01:55	durka42	and calls it for every pixel
01:56	durka42	which is a bad design IMO
01:56	hiredman	haha
01:56	ayrnieu	yes, that's horrid.
01:56	durka42	i mean, i don't know if i'd want to call even a C function (* 320 240) times for each frame that comes in from theh camera
01:56	eee	now I see why you want to do it in parallel
01:57	ayrnieu	Jesus, it's like they never heard of APL.
01:57	durka42	i think i can get the pixel array
01:57	eee	gnight'
01:57	durka42	night
01:59	durka42	does anyone know what's up with clojure.parallel, though?
01:59	durka42	where might i find jsr166y.forkjoin.ParallelArray
01:59	hiredman	clojurebot: google jsr166y
01:59	clojurebot	First, out of 725 results is:
01:59	clojurebot	jsr166y overview
01:59	clojurebot	http://artisans-serverintellect-com.si-eioswww6.com/default.asp?W9
01:59	ayrnieu	./src/clj/clojure/parallel.clj:You'll need jsr166y.jar in your classpath in order to use this
02:00	ayrnieu	http://gee.cs.oswego.edu/dl/concurrency-interest/index.html , there's a big comment about it in that source.
02:00	durka42	thanks
02:00	durka42	i had just searched for the classname and nothing useful came up
02:01	ayrnieu	I do a lot of `find . \|xargs egrep`
02:01	hiredman	erm
02:01	hiredman	-R
02:02	ayrnieu	find .\|egrep clj\|xargs egrep
02:02	hiredman	ugh
02:02	hiredman	find . -name \*.clj
02:02	ayrnieu	find .\|ergep -v git\|xargs egrep
02:03	hiredman	find . -name \.clj -not -name \git\*
02:03	ayrnieu	Yeah, I'll remember that for another few minutes.
02:03	hiredman	find . -name \.clj -not -name \git\* -exec egrep Foo {} \;
02:04	hiredman	find is one of my favorite tools
02:05	ayrnieu	so you're a Java man at the shell and a haskell man at the editor :-)
02:05	hiredman	eh?
02:10	durka42	oh no!
02:10	durka42	QTJava requires java 5
02:10	durka42	clojure.parallel require java 6
02:10	hiredman	ouch
02:11	durka42	maybe i will have to try that 32-bit java 6 build
02:11	durka42	but QTJava is obsolete anyway
02:11	durka42	unfortunately there doesn't seem to be a working java interface to the new way
02:11	durka42	hmm, that wasn't the problem
02:12	durka42	jsr166 doesn't quite compile
02:12	durka42	oh there is a binary
02:17	durka42	i think ParallelArray moved...
02:18	durka42	jsr166y.forkjoin.ParallelArray --> extra166y.ParallelArray
02:18	durka42	commit message from 1/6: Repackaged parallel collection classes
02:19	hiredman	hmmm
02:20	durka42	things work if i go into parallel.clj and swap out jsr166y.forkjoin for extra166y at the top
02:21	durka42	user=> (reduce + 0 [1 2 3 4 5])
02:21	durka42	15
02:21	durka42	user=> (preduce + 0 [1 2 3 4 5])
02:21	durka42	15
02:21	cp2	postduce?
02:22	ozy`	,(doc preduce)
02:22	clojurebot	java.lang.Exception: Unable to resolve var: preduce in this context
02:22	durka42	,(use 'clojure.parallel)
02:22	clojurebot	java.io.FileNotFoundException: Could not locate clojure/parallel__init.class or clojure/parallel.clj on classpath:
02:22	ozy`	wait, parallel reduce, right?
02:22	durka42	yes
02:24	durka42	range is optimized for reduce, yes? but if i call seq on it, does that remove the optimizations?
02:27	durka42	should i open an issue for the moved imports thing?
02:28	LordOfTheNoobs	Do you actually have multiple cores? :) Also, is this for large or small data sets? I think that thread spawn would get you for smaller data sets.
02:29	durka42	yes, sort of, can you elaborate?
02:29	durka42	this was my test: (time (dotimes [_ 1000] (doall (pmap inc (seq (range 1000))))))
02:29	durka42	with that example pmap is 24 times slower than map
02:33	ayrnieu	,(first (lazy-cons (print 1) (print 2) (print 3)))
02:33	clojurebot	1
02:33	LordOfTheNoobs	Do not take me as an authority, for I am on the other end of the knowledge pool here. For only a thousand items with a fast computation on each the amount of time it take the JVM to setup and dispatch to the subthreads may outweight the entire runtime of a single buteforce thread. Doing things in parallel tends to win when you have many cores, a hard computation, and many items to apply it to.
02:33	ayrnieu	,(first (seq (lazy-cons (print 1) (print 2) (print 3))))
02:34	clojurebot	1
02:34	ayrnieu	,(rest (seq (lazy-cons (print 1) (print 2) (print 3))))
02:34	clojurebot	123
02:34	durka42	i believe that
02:34	durka42	i upped it to 100,000 items, which should take a while
02:34	durka42	well, not "should"
02:34	durka42	but is
02:35	durka42	i think i only have two cores, so it won't speed up much
02:37	LordOfTheNoobs	Good luck with your benchmark durka. I'd re-ask tommorow for a set of likely more precise and accurate answers.
03:12	Lau_of_DK	Bonjour cgrand ! :)
03:14	Raynes	Hai.
03:15	cgrand	Hej Lau!
03:20	hiredman	purely functional then?
03:23	Raynes	No, it's an overcomplicated hello-world program :p
03:23	Raynes	Still useless.
04:11	Lau_of_DK	Raynes :)
04:11	Raynes	Lau_of_DK: Yeah I know, I've been in here for a month and never actually used Clojure. ._.
04:12	Raynes	Until now.
04:12	Raynes	:D
04:20	Lau_of_DK	Raynes, sorry I left out "hi" I wasnt making a comment :)
04:20	Raynes	I realized.
04:21	Lau_of_DK	Good :) Just being friendly
05:24	BigTom	kotarak: Hi
05:24	kotarak	Salut.
05:25	BigTom	my GUI is not fixed but I have made progress
05:26	kotarak	Good to hear. :)
05:26	BigTom	The outstanding issue is that my watcher seems to disappear after its first call
05:27	BigTom	(The RejectedExecutionException was a red-herring, I had shut down agents in the repl)
05:27	kotarak	:P
05:28	BigTom	Doh!
05:28	kotarak	Sh** happens. ;)
05:28	BigTom	If I update the GUI from the watcher it gets called once then vanishes
05:29	BigTom	If I replace the gui update with a prn it keeps getting updates
05:29	BigTom	Its odd (but I, suspect, fundamentally dumb on my part)
05:34	kotarak	"Blame yourself first" helps you to fix 98% of all the bugs.
05:34	kotarak	In particular: it's almost never a compiler bug....
05:38	Chousuke	hmm :/
06:12	zakwilson	I'm trying to make a gui update a counter showing completion of an operation being applied to a bunch of files with pmap. Getting the counter to work is easy using either an agent or a ref, but the UI updes rarely and gets very out of sync.
06:13	zakwilson	I've tried putting a JTextField in a ref or an agent and did not get better results.
06:13	BigTom	Is there any way you can see the watchers attached to an agent?
06:14	BigTom	zakwilson: have you looked at cells?
06:15	turbo24prg	morning
06:15	zakwilson	No.
06:16	BigTom	zakwilson: http://github.com/timothypratley/strive/blob/8285ef1419601411797205de0d60a2b2c0b71be0/src/clj/cells.clj
06:19	turbo24prg	i'd like to build a datastructure based on nested sorted-maps. i'm familiar with common lisp, but new to clojure. what is the best method to modify it? refs/transactions, agents or atoms?
06:20	kotarak	turbo24prg: assoc-in and update-in
06:21	BigTom	turbo24prg: It really depends on the level of coordination required between the threads
06:22	turbo24prg	but that wouldn't be save if i'd have to update several nested maps in there as one transaction, and would like to read from it
06:23	turbo24prg	for one action i'd have to perform several inserts/updates on different maps in the structure, several levels deep
06:23	kotarak	sounds like a ref in combination with assoc-in/update-in.
06:24	turbo24prg	during update i'd like to have read-access on a stable version of it. isn't that ref/transactions?
06:25	BigTom	you cannot update a ref outside a transaction
06:25	BigTom	so if you use a ref you are going to use transactions :-)
06:25	turbo24prg	the example for refs on the website isn't the easiest for a newbie
06:26	turbo24prg	especially if you're not into java's concurrency
06:27	kotarak	turbo24prg: refs are easy. (def a-ref (ref {:a {:x 5} :b {:y 6}})) (dosync (alter a-ref assoc-in [:a :x] 7)) => {:a {:x 7} :b {:y 6}}
06:28	kotarak	The code outside the dosync always sees a stable version.
06:28	turbo24prg	uh, that's a nice example.
06:28	turbo24prg	thanks!
06:29	BigTom	also, have a look on http://en.wikibooks.org/wiki/Clojure_Programming/By_Example
06:29	kotarak	or with update: (dosync (alter a-ref update-in [:b :y] inc)) => {:a {:x 7} :b {:y 7}}
06:30	kotarak	A lot of things in Clojure work similar. So when you grok the idea once, it's easy to transfer to atoms, agents, etc...
06:32	turbo24prg	concurrency is sth. i really missed in common lisp
06:32	turbo24prg	very cool
06:33	eleftherios	what's is the most mature web framework for clijure so far? Compojure?
06:33	turbo24prg	did you came from another lisp? or are you "java guys"?
06:34	kotarak	turbo24prg: I'm more a Scheme guy. Having more experience in C and a bit OCaml.
06:35	Raynes	I come from Haskell. With a little experience in C# and several other irrelevant languages.
06:35	BigTom	turbo24prg: while I'm a Java guy (which why I keep asking the guys here for help!)
06:35	turbo24prg	wow.
06:36	kotarak	Good mixture. :)
06:36	turbo24prg	yeah :)
06:36	BigTom	turbo24prg: clojure is definitely a Lisp, so you should feel very at home
06:37	zakwilson	BigTom: waching-label looks cool, but it did not work for me. The agent updates in real-time, but the GUI does not.
06:37	turbo24prg	nice, assoc supports "paths"
06:37	kotarak	Clojure has everything, which I missed about other Lisps: Immediate access to libraries (Java, but hey, it gets the work done) and the "smallness" of Scheme (in terms of core functions)
06:38	kotarak	turbo24prg: there is also a "normal" assoc
06:38	BigTom	zakwilson: Hm, I 'm having problems with the pattern too
06:38	BigTom	zakwilson: does cells itself work for you?
06:38	zakwilson	Using watchers has the same results I was having trying to do it manually.
06:39	zakwilson	I haven't tried cells itself - just watching-label.
06:39	BigTom	zakwilson: try running cells to confirm your environment is ok
06:41	zakwilson	I just did. It works.
06:43	zakwilson	I just tried using an atom instead of an agent since that's what cells does. No change.
06:43	BigTom	zakwilson: could you post your code?
06:44	BigTom	lisppaste8: url
06:44	lisppaste8	To use the lisppaste bot, visit http://paste.lisp.org/new/clojure and enter your paste.
06:46	zakwilson	I'm not sure how much I can separate the problem code from the whole app and still reproduce the problem.
06:46	zakwilson	And the whole app is a bit big for a paste...
06:46	zakwilson	So... http://github.com/zakwilson/imagesieve/tree/master
06:47	zakwilson	This is before any attempt to use a watcher, and the UI code is a bit more messy than I'd like.
06:48	zakwilson	Wait... does this thing depend on my utilities file? I need to check.
06:49	zakwilson	It does. I'll put that up on github too.
06:51	zakwilson	Dependency: http://github.com/zakwilson/zutil-clj/tree/master
06:51	cads	what's the -> macro called?
06:52	zakwilson	I think I would call it "arrow".
06:53	cads	it's pretty funky :D
06:55	cads	takes a list of forms headed by a value, evaluates each form with the result from the previous form inserted into the second position of the form
06:55	cads	I think.
06:55	cads	Do we use it often?
06:58	cads	if we have a few simple one parameter functions we can use it to compose function application like (-> x (f1) (f2) (f3)) => (f3 (f2 (f1 x)))
07:10	BigTom	zakwilson: just running it up now...
07:11	Chousuke	cads: it has many uses :)
07:11	Chousuke	cads: you can omit the parens though I think
07:11	zakwilson	BigTom: (update-counter) works fine when it's not actively processing.
07:11	Chousuke	,(macroexpand-1 '(-> f1 f2 f3))
07:11	clojurebot	(clojure.core/-> (clojure.core/-> f1 f2) f3)
07:12	Chousuke	yeah, seems to work :/
07:13	Chousuke	,(macroexpand '(-> f1 f2 f3))
07:13	clojurebot	(f3 (clojure.core/-> f1 f2))
07:15	BigTom	zakwilson: well, I have some gray jpegs now :-)
07:16	zakwilson	But no up-to-date GUI saying how many.
07:19	BigTom	zakwilson: indeed
07:22	BigTom	zackwilson: it updates when you mouse-over the GUI
07:22	zakwilson	BigTom: I may have improved things a bit. I moved the processing itself in to an agent.
07:23	zakwilson	The update isn't quite in sync that way, but it's pretty close.
07:36	BigTom	actually, it does update
07:37	turbo24prg	is there any way to differentiate between different types of struct?
07:37	cads	can we write a macro that takes a form with arithmetic written with infix syntax, such as ((b * b) - 4 * a * c), and turns it the well formatted lispy equivalent, (- (* b b) (* 4 a c))? would this be considered a perversion?
07:37	turbo24prg	say (defstruct a ..) (defstruct b ...). in CL i'd use typecase
07:38	BigTom	zakwilson: For me it updates the GUI pretty much at the same time the file appears
07:41	zakwilson	BigTom: What sort of hardware are you running it on?
07:43	BigTom	a single core ancient ubuntu box
07:44	zakwilson	Interesting. There is a small logical difference in how it does things on single-core machines (map instead of pmap).
07:44	zakwilson	I'm about to try it on an older single-core Windows box.
07:49	Chousuke	cads: sure it's possible; just not trivial :)
07:51	Chousuke	personally I don't think the infix syntax offers anything over prefix syntax :/
07:53	Chousuke	unless you're going to allow writing things like (b^2 - 4ac), but that's even more complicated.
07:53	cads	unless you're doing algebra in ascii, I guess
07:54	cads	that's not so bad, your stuff is already in tokens, you just got to do a little parsing :D
07:54	Chousuke	but in essence, you "only" need a function that takes a list of operators and operands in infix form and transforms it into a similar list in prefix form :)
07:55	cads	I cant really think of examples of algebra where prefix notation is absolutely horrific
07:56	Chousuke	it's maybe not the easiest to read if you put it all on one line.
07:57	cads	I can think of places where infix operators would be more mnemonic and easier to parse for a human reader
07:57	Chousuke	something like (/ (* 5 (- 4 z)) 2)
07:57	zakwilson	BigTom: The latest version works as intended everywhere, without mousing over the UI. I think the problem was the async update to the GUI by a callback from the GUI's own thread.
07:57	Chousuke	but that's really much better if it's properly indented.
07:58	Chousuke	lisppaste8: url
07:58	lisppaste8	To use the lisppaste bot, visit http://paste.lisp.org/new/clojure and enter your paste.
07:59	lisppaste8	Chousuke pasted "like this" at http://paste.lisp.org/display/74634
07:59	cads	eh, that's not too bad, I bet one gets used to it
07:59	cads	looking at your second example now
08:00	Chousuke	that even looks a lot like how you'd write it if you were writing using real notation instead of ascii formulas
08:00	cads	yeah, with / you can separate the numerator from the denominators and it looks nice :)
08:27	turbo24prg	assoc supports paths, but find doesn't? :/
08:31	turbo24prg	assoc-in supports them, find/get doesn't.
08:39	Chousuke	get-in? :)
08:39	Chousuke	,(doc get-in)
08:39	clojurebot	"([m ks]); returns the value in a nested associative structure, where ks is a sequence of keys"
08:41	Chousuke	even works with vectors if you supply integer keys :)
08:41	BigTom	zakwilson: cool
08:54	turbo24prg	uh, cool
08:59	kotarak	Hmmm.. Should get-in support a default value?
09:02	BigTom	,(includes? :k {:k :v :k1 :v})
09:02	clojurebot	java.lang.Exception: Unable to resolve symbol: includes? in this context
09:02	BigTom	oops
09:02	kotarak	(doc contains?)
09:02	clojurebot	Returns true if key is present in the given collection, otherwise returns false. Note that for numerically indexed collections like vectors and Java arrays, this tests if the numeric key is within the range of indexes. 'contains?' operates constant or logarithmic time; it will not perform a linear search for a value. See also 'some'.; arglists ([coll key])
09:03	BigTom	kotarac: cheers
09:05	kotarak	BigTom: another useful repl function
09:05	kotarak	(doc find-doc)
09:05	clojurebot	Prints documentation for any var whose documentation or name contains a match for re-string-or-pattern; arglists ([re-string-or-pattern])
09:06	BigTom	kotarak: I can never remember that one
09:07	BigTom	Its very weird, I had an app that worked 3 weeks back and its busted all over now
09:07	BigTom	I think I must have every one of the bad idioms Rich is tidying up :-(
09:08	Hun	kotarak: though i liked the old name apropos more :)
09:08	hoeck	were there any breaking changes in the last 3 weeks??
09:09	BigTom	I don't know, watchers changed in early Jan
09:09	BigTom	I used clojure-contrib includes? when I should have used contains - see above
09:10	BigTom	So, its all down to me really :-/
09:11	hoeck	ah, okay
09:11	BigTom	kotarak: I fixed the watcher thing, I had too many braces
09:11	kotarak	Hun: (def apropos find-doc) ;)
09:11	Hun	:)
09:12	Hun	of course. but for some reason, i found find-doc unintuitive
09:12	Hun	though that might be just musclememory to hit (apropos " before thinking
09:12	kotarak	I find it mnemonic.
09:13	kotarak	apropos is used here a bit differently.
09:19	BigTom	agent question
09:19	BigTom	if I send actions to an agent from one thread then is the execution order guaranteed to be the same as the send order?
09:19	Chouser	BigTom: yes
09:20	ayrnieu	one of the first things I wrote in clojure: http://paste.lisp.org/display/74207 - apropos. It needs some work.
09:20	BigTom	thought so
09:20	BigTom	thanks
09:21	Chouser	Ah, here it is, tucked into the middle of a paragraph: "Actions dispatched to an agent from another single agent or thread will occur in the order they were sent, potentially interleaved with actions dispatched to the same agent from other sources."
09:21	Chouser	from http://clojure.org/agents
09:22	ayrnieu	...which explains how an agent 'looping' over sends to itself can be stopped.
09:23	BigTom	Chouser: Thank you for explaining before RTFMing :-)
10:58	turbo24prg	hmm, (doseq [a ... b ...] ... does loop a1 b1, a1 b2, ...
10:58	turbo24prg	thought about (doseq [[a b] (interleave as bs)] ... but it doesn't work because interleave creates no sublists
10:59	turbo24prg	any handy way to iterate over multiple lists element for element?
10:59	kotarak	(doc partition)
10:59	clojurebot	Returns a lazy sequence of lists of n items each, at offsets step apart. If step is not supplied, defaults to n, i.e. the partitions do not overlap.; arglists ([n coll] [n step coll])
10:59	kotarak	(partition 2 (interleave a b))
10:59	turbo24prg	neat
10:59	turbo24prg	thanks!
11:02	kotarak	For two things, one could maybe exploit zipmap. But that would be an ugly hack.
11:09	Chouser	what about just (map ... a b) ?
11:10	kotarak	doseq suggest side effects. So (dorun (map ... a b)), beh. :/
11:13	Chouser	mm, good point.
11:19	BigTom	how would I get a lazy sequence of random numbers?
11:20	ayrnieu	,(take 30 (repeatedly #(rand-int 30)))
11:20	clojurebot	(15 12 3 12 2 10 7 23 4 2 22 13 17 9 19 14 10 8 6 4 1 4 27 17 6 9 11 22 2 5)
11:20	ayrnieu	,(.contains (apply str (take 1000000 (repeatedly #(char (+ (int \a) (rand-int 26)))))) "ayrnieu")
11:20	thickey	rhickey or Chouser: where can i find the script that currently generates the api page?
11:20	clojurebot	false
11:22	BigTom	ayrnieu: thanks
11:29	Chouser	thickey: I've never seen that script, only heard rumors of its existance.
11:30	thickey	Chouser: =) okay, thanks
11:37	BigTom	ok so I have this:
11:37	BigTom	,(into #{} (partition 2 (interleave [1 2 5 4] [5 6 1 8])))
11:37	clojurebot	#{(2 6) (4 8) (1 5) (5 1)}
11:38	BigTom	but I really want a set of vectors
11:38	ayrnieu	,(class #{(2 6})
11:38	clojurebot	Eval-in-box threw an exception:Unmatched delimiter: }
11:38	ayrnieu	OK. (map vector ...) ?
11:39	ayrnieu	well, I'd think that'd work, but it doesn't.
11:39	ayrnieu	,(into #{} (map #(apply vector %) (partition 2 (interleave [1 2 3 4] [5 6 1 8]))))
11:39	clojurebot	#{[2 6] [4 8] [1 5] [3 1]}
11:39	kotarak	(doc vec)
11:39	clojurebot	Creates a new vector containing the contents of coll.; arglists ([coll])
11:40	kotarak	,(map vec (partition 2 (interleave [1 2 3] [4 5 6])))
11:40	clojurebot	([1 4] [2 5] [3 6])
11:40	BigTom	ah, sweet
11:40	ayrnieu	just found that with (find-doc "vector") :-)
11:46	BigTom	another includes? bites th edust
11:46	thickey	Chouser: this looks like it: http://paste.lisp.org/display/71857
11:49	Chouser	thickey: ah, thanks.
11:51	thickey	i get a java.lang.ClassNotFoundException: jsr166y.forkjoin.ParallelArray when running it. i know i've had that issue before but forget how to sort it out
11:52	ayrnieu	http://gee.cs.oswego.edu/dl/concurrency-interest/index.html , but other changes are needed.
11:52	ayrnieu	durka42 went through it earlier: http://clojure-log.n01se.net/
11:54	Chouser	,(map vector [1 2 3] [4 5 6])
11:54	clojurebot	([1 4] [2 5] [3 6])
12:19	thickey	rhickey: does the jsr166y.jar that parallel.clj was developed with no longer match the one located here? http://gee.cs.oswego.edu/dl/concurrency-interest/index.html
12:20	rhickey	thickey: there's a jsr166y.jar in the files section of the google group - try that
12:20	thickey	rhickey: it seems like there was some refactoring and the parallel classes are now in extra166y.jar
12:20	thickey	rhickey: ok
12:24	turbo24prg	to add an element to a sorted-set i need to (apply sorted-set (cons 3 (sorted-set 5 1 3 2 4 1)) to get back a sorted-set again, right?
12:24	kotarak	(conj the-set the-thing)
12:25	turbo24prg	neat. thanks. finding all the methods is hard. and it's cool if they exist :)
12:26	kotarak	conj works also for maps, vectors, lists, ...
12:28	Vargr2	I can't seem to get sensible error messages and tab-completion to work with the current versions of clojure (trunk, HEAD), clojure-mode (HEAD) and swank-clojure (HEAD) - is that a known problem or am I missing something?
12:28	turbo24prg	CL felt a lot more practical then scheme, but clojure seems even more practical
12:30	turbo24prg	Vargr2: updated this morning, seems to work here
12:31	Vargr2	turbo24prg: I get "Synchronous Lisp Evaluation aborted" when trying to expand. No problem with CLISP and SBCL, so SLIME should be alright, no?
12:31	kotarak	turbo24prg: http://clojure.org/api gives a nice summary
12:31	kotarak	for functions I mean
12:34	turbo24prg	kotarak: i'm browsing /api, /sequences all the time :)
12:37	turbo24prg	Vargr2: i'm using 23.0.60.1, no probs. duno
12:39	turbo24prg	it shouldn't be possible that the hash-codes of strings would < but the strings themself >, right?
12:39	Chousuke	why not? :/
12:39	Chousuke	I don't think hash codes support ordering
12:41	turbo24prg	http://java.sun.com/j2se/1.4.2/docs/api/java/lang/String.html#hashCode()
12:42	turbo24prg	,(hash "6")
12:42	clojurebot	54
12:42	turbo24prg	,(hash "7")
12:42	clojurebot	55
12:42	turbo24prg	,(hash "71")
12:42	clojurebot	1754
12:42	durka42	,(hash "7")
12:42	clojurebot	55
12:43	turbo24prg	imho they're ordered, no?
12:44	ayrnieu	,(hash "merry had a little lamb whose fleece was white as snow, and every where that mary went the lamb was sure to go")
12:44	clojurebot	-1376680459
12:44	ayrnieu	it's positive if you spell her name correctly, but still a counterexample.
12:45	clojurebot	svn rev 1241; fixed pmap so parallel in first n steps
12:46	turbo24prg	arbscht: thanks
12:46	turbo24prg	erm, ayrnieu.
12:48	durka42	,(Integer/toHexString -1376680459)
12:48	clojurebot	"adf185f5"
12:50	turbo24prg	so, to compare strings i'd have to call java then?
12:51	turbo24prg	no string<, string= ...
12:55	Chouser	,(compare "abc" "abc")
12:56	Chouser	(doc compare)
12:56	Chouser	clojurebot's dead
12:58	durka42	you can't just use = ?
12:58	durka42	= uses .equals internally
13:00	turbo24prg	but < works only on numbers
13:01	durka42	oh, true
13:02	durka42	(doc compare)
13:02	clojurebot	Comparator. Returns 0 if x equals y, -1 if x is logically 'less than' y, else 1. Same as Java x.compareTo(y) except it also works for nil, and compares numbers and collections in a type-independent manner. x must implement Comparable; arglists ([x y])
13:02	durka42	(compare "bar" "bar")
13:02	durka42	,(compare "bar" "bar")
13:02	clojurebot	0
13:02	durka42	,(compare "bar" "quux")
13:02	clojurebot	-15
13:02	durka42	,(compare "quux" "bar")
13:03	clojurebot	15
13:05	ayrnieu	,(let [m "merry had a little lamb whose fleece was white as snow, and every where that mary went the lamb was sure to go"] [(hash m) (sort [(hash m) (hash "a")]) (sort [m "a"])])
13:05	clojurebot	[-1376680459 (-1376680459 97) ("a" "merry had a little lamb whose fleece was white as snow, and every where that mary went the lamb was sure to go")]
13:07	ayrnieu	but if you're paranoid, (sort #(.compareTo %1 %2) ...)
13:07	turbo24prg	uh, i still thing a way to cl. cool that there are so many generic functions
13:07	turbo24prg	thanks.
13:51	clojurebot	svn rev 1242; [lazy] switched concat to lazy-seq, added Sequence, printing of sequences, more returns non-nil
13:54	adakkak	can someone tell me what is wrong with this basic program? I am just learning clojure http://paste.lisp.org/display/74642
13:55	leafw	adakkak: ([] ([])) makes little sense. You meant ([] []) ... yet it still makes little sense: for no args, return empty vector?
13:56	adakkak	leafw: what should I be returning?
13:56	Chousuke	adakkak: that looks like you're trying to do haskell-style pattern matching
13:57	Chousuke	clojure doesn't support it :/
13:57	adakkak	does clojure have any pattern matching for function's arguments?
13:58	Chousuke	no.
13:58	Chousuke	well, there's destructuring
13:58	Chousuke	which is kind of similar, but not pattern matching.
13:59	Chousuke	,(apply str (map #(str % "s") '("this" "is" "a" "test")))
13:59	clojurebot	"thississastests"
13:59	Chousuke	hmm
13:59	Chousuke	not quite
13:59	adakkak	do you know if scheme has pattern matching?
13:59	Chousuke	I'm not sure, but I don't think it does.
14:00	Chousuke	adakkak: what you're doing in that piece of clojure code is defining two overloads of the same function, just of different arity.
14:01	Chousuke	you're never even calling the nullary version :)
14:01	eyeris	Could someone point me to the explaination of the & symbol in defmacros?
14:01	adakkak	Chousuke: so why doesn't the piece of code work?
14:01	Chousuke	adakkak: you don't have a termination clause. it keeps recursing infinitely.
14:03	Chousuke	at some point it'll call "(plural-sentence nil)" but that's still calling the one-argument version, which will again call (plural-sentence nil)
14:03	Chousuke	since (rest nil) is nil
14:03	adakkak	Chousuke:Ahh, so (rest ()) is nil?
14:03	Chousuke	yeah; though you'll never see a () there :/
14:03	Chousuke	nil is not equal to ()
14:04	Chousuke	,(rest nil)
14:04	clojurebot	nil
14:04	Chousuke	,(rest '())
14:04	clojurebot	nil
14:04	Chousuke	(= '() nil)
14:04	Chousuke	,(= '() nil)
14:04	clojurebot	false
14:05	Chousuke	,(rest '(1))
14:05	clojurebot	nil
14:10	adakkak	Chousuke:I fixed my little program and am now wondering why I get ("test")s in the output for http://paste.lisp.org/display/74642#2
14:13	ayrnieu	adakkak: (if (= (count sentence) 1) (plural_word sentence)
14:14	adakkak	ayrnieu: what is wrong with it?
14:15	Chousuke	adakkak: http://paste.lisp.org/display/74642#3
14:15	Chousuke	adakkak: the sentence is still a list
14:16	Chousuke	adakkak: you want the word from it
14:16	Chousuke	,(str (first '("test")))
14:16	clojurebot	"test"
14:17	Chousuke	if you forget the first, you get the parens too
14:17	Chousuke	adakkak: as an aside, use dashes instead of underscores in naming
14:17	Chousuke	plural-word instead of plural_word
14:17	adakkak	Chousuke: is that the convention in clojure?
14:18	ayrnieu	that's the convention in every language that permits dashes in symbols.
14:18	dreish	adakkak: It is in all Lisps, but beyond convention, I think underscores in Clojure are informally reserved for gensyms.
14:19	adakkak	I am wondering if having (plural-word (first sentence)) twice in the code is considered bad practice in clojure?
14:20	Chousuke	it's not a deadly sin
14:21	adakkak	also, how do you know that there is a keyword called remaining? I am checking the api and there is no mention of it!
14:25	Chousuke	adakkak: it's not a keyword
14:25	Chousuke	adakkak: I bind it, in the if-let
14:25	aartist_	why clojure?
14:26	Chousuke	adakkak: http://paste.lisp.org/display/74642#5
14:26	adakkak	so that is the difference between a regular if and if-let
14:26	Chousuke	adakkak: if-let makes a binding at the same time
14:26	adakkak	makes sense
14:27	Chousuke	it's a shortcut for (let [foo bar] (if foo dostuff do-otherstuff))
14:27	Chousuke	ie. that would be: (if-let [foo bar] dostuff do-otherstuff)
14:28	Chousuke	where the do-otherstuff is executed when "foo" is non-true
14:29	keithb	Is it true that, in a given dosync transaction, all mutations will occur in sequential order, even if they are commute's and not alter's?
14:30	Chousuke	"sequential order"?
14:30	keithb	I mean in the order in which they were specified in the code.
14:30	Chousuke	I think with commutes the order is not guaranteed
14:30	adakkak	Chousuke: how does foo become false, I though even nil does not evaluate to false
14:31	Chousuke	adakkak: hm? :/
14:31	Chousuke	both "nil" and "false" are non-true values in clojure
14:31	ayrnieu	http://paste.lisp.org/display/74642#6 defmulti version.
14:32	Chousuke	,(if nil "nil is true" "nil is false")
14:32	clojurebot	"nil is false"
14:32	Chousuke	(if-let [foo nil] :true :false)
14:32	Chousuke	gah
14:33	Chousuke	,(if-let [foo nil] [:true foo] [:false foo])
14:33	clojurebot	java.lang.Exception: Unable to resolve symbol: foo in this context
14:33	Chousuke	... okay, I fail.
14:33	Chousuke	foo is of course not bound if it'd be false :P
14:33	ayrnieu	,(if-let [foo true] [:true foo] [:false foo])
14:33	clojurebot	java.lang.Exception: Unable to resolve symbol: foo in this context
14:34	Chousuke	,(if-let [foo 1] foo :nothing)
14:34	clojurebot	1
14:34	Chousuke	,(if-let [foo nil] foo :nothing)
14:34	clojurebot	:nothing
14:34	ayrnieu	sensible.
14:37	adakkak	Chousuke: got it
14:53	lisppaste8	ayrnieu pasted "defmulti examples" at http://paste.lisp.org/display/74647
14:53	ayrnieu	translated from mercury translated from haskell.
14:55	ayrnieu	I backquoted json-ex with the idea of showing that you can use symbols naturally and still execute code, but defmulti doesn't take kindly to symbols. Oh well.
15:23	hiredman	ayrnieu: I imagine you wanted ' not `
15:23	ayrnieu	hiredman - no, I want ` there.
15:24	hiredman	ok
15:24	Chousuke	ayrnieu: maybe defmulti didn't like the namespace-qualified symbols.
15:24	Chousuke	,(= 'foo `foo)
15:24	clojurebot	false
15:25	hiredman	,`foo
15:25	clojurebot	clojure.core/foo
15:25	hiredman	,'foo
15:25	clojurebot	foo
15:26	hiredman	huh
15:26	Chousuke	hiredman: did you notice my (doc fix? (I hope I didn't break anything)
15:26	hiredman	Chousuke: no I did not notice anything
15:27	ayrnieu	aha. OK, defmulti does work with symbols, but his `foo 'foo difference only reemphasises that symbols will not work for data-structures like this.
15:32	hiredman	(doc namespace)
15:32	clojurebot	Returns the namespace String of a symbol or keyword or nil if not present ; arglists ([x])
15:32	hiredman	(doc for)
15:32	clojurebot	List comprehension Takes a vector of one or more binding form collection expr pairs each followed by an optional filtering when while expression when test or while test and yields a lazy sequence of evaluations of expr Collections are iterated in a nested fashion rightmost fastest and nested coll exprs can refer to bindings created in prior binding forms take 100 for x range 100000000 y range 1000000 while y x x y ; argli
15:32	hiredman	excellent
15:34	Chousuke	the docstrings aren't properly punctuated, which is a slight problem
15:37	cooldude127	Chousuke: what's changed?
15:37	Chousuke	it removes whitespace and handles nil arglists
15:38	Chousuke	(doc print-dup)
15:38	clojurebot	When set to logical true objects will be printed in a way that preserves their type when read in later Defaults to false
15:38	Chousuke	no arglist :)
15:39	cooldude127	oh
15:39	cooldude127	but why does it remove punctuation?
15:39	Chousuke	it doesn't.
15:40	Chousuke	there's no punctuation to remove :P
15:40	cooldude127	oh. i'm confused
15:40	hiredman	Chousuke: maybe [^ ] instead of \\W or whatever
15:40	Chousuke	see the docstring for print-dup in your repl.
15:41	cooldude127	Chousuke: i see nil
15:41	cooldude127	maybe i'm behind
15:41	Chousuke	heh, yeah
15:41	cooldude127	Chousuke: but my (doc for) has periods
15:41	hiredman	(doc for)
15:41	clojurebot	; arglists ([seq-exprs expr])
15:41	Chousuke	I guess I forgot proper punctiation from teh docstrings
15:41	hiredman	whoops
15:42	Chousuke	:D
15:42	cooldude127	lol
15:42	hiredman	(doc for)
15:42	clojurebot	List comprehension Takes a vector of one or more binding form collection expr pairs each followed by an optional filtering when while expression when test or while test and yields a lazy sequence of evaluations of expr Collections are iterated in a nested fashion rightmost fastest and nested coll exprs can refer to bindings created in prior binding forms take 100 for x range 100000000 y range 1000000 while y x x y ; argli
15:43	cooldude127	see even the parens and stuff for the example are gone, that can't be good
15:45	hiredman	(doc for)
15:45	clojurebot	List comprehension. Takes a vector of one or more binding-form/collection-expr pairs, each followed by an optional filtering :when/:while expression (:when test or :while test), and yields a lazy sequence of evaluations of expr. Collections are iterated in a nested fashion, rightmost fastest, and nested coll-exprs can refer to bindings created in prior binding-forms. (take 100 (for [x (range 100000000) y (range 1000000) :
15:45	cooldude127	yay
15:46	hiredman	switched from \\W+ to \\s+
15:47	MarkVolkmann	The REPL doesn't recognize the union function.
15:48	MarkVolkmann	(doc union) comes back with "Unable to resolve var: union in this context".
15:48	MarkVolkmann	I'm trying to use it to combine two sets into one.
15:48	hoeck	,(doc clojure.set/union)
15:48	clojurebot	"([xset yset]); Returns a set that is the union of the two sets."
15:48	Chousuke	MarkVolkmann: it's not in the clojure.core namespace :)
15:49	MarkVolkmann	Thanks! Looking at http://clojure.org/api, I'm not sure how I could have known that.
15:49	Chousuke	hmm
15:50	hiredman	http://clojure.org/api#toc588 <-- it is in the clojure.set section of the api
15:50	Chousuke	it does have the section titled
15:50	Chousuke	but it doesn't really stand out.
15:50	MarkVolkmann	Ah, I see. It's kind of hidden, just after the documentation for pvec.
15:50	MarkVolkmann	It's in the same font as all the function names, so it's easy to miss.
15:50	Chousuke	definitely needs improvement :/
15:51	Chousuke	even better would be to have everything explicitly show their namespace.
15:54	ayrnieu	,(name 'user/x)
15:54	clojurebot	"x"
15:54	ayrnieu	how do I get "user" out of that?
15:54	ayrnieu	,(namespace 'user/x) ;; oh.
15:54	clojurebot	"user"
15:56	ozy`	,(map + (range 20) (range 0 40 2))
15:56	clojurebot	(0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57)
15:56	ozy`	,(map #'list (range 20) (range 0 40 2))
15:56	clojurebot	((0 0) (1 2) (2 4) (3 6) (4 8) (5 10) (6 12) (7 14) (8 16) (9 18) (10 20) (11 22) (12 24) (13 26) (14 28) (15 30) (16 32) (17 34) (18 36) (19 38))
15:57	hiredman	,(map list (range 20) (range 0 40 2))
15:57	clojurebot	((0 0) (1 2) (2 4) (3 6) (4 8) (5 10) (6 12) (7 14) (8 16) (9 18) (10 20) (11 22) (12 24) (13 26) (14 28) (15 30) (16 32) (17 34) (18 36) (19 38))
15:58	ozy`	,;
15:58	clojurebot	Eval-in-box threw an exception:EOF while reading
15:58	ozy`	,(display 'woot)
15:58	clojurebot	java.lang.Exception: Unable to resolve symbol: display in this context
15:58	ozy`	,(println '(pizza))
15:58	clojurebot	(pizza)
16:01	ozy`	,(println '(SYMBOLS IN ALL CAPS))
16:01	clojurebot	(SYMBOLS IN ALL CAPS)
16:15	ayrnieu	http://gist.github.com/55662 <-- similar, but checks at compile-time that you've handled every case.
16:19	hiredman	clojurebot: parens is <reply>????? ????
16:19	clojurebot	Roger.
16:20	Chousuke	?)
16:21	hiredman	oooh
16:22	hiredman	google translate has a "ajax" (json) api
16:22	danlarkin	I smell a new feature
16:56	blbrown	does anyone remember where the clojure api examples are on the wiki
16:57	blbrown	nevermind, http://en.wikibooks.org/wiki/Clojure_Programming/Examples/API_Examples
16:57	LordOfTheNoobs	clojurebot: api examples?
16:57	clojurebot	examples is http://en.wikibooks.org/wiki/Clojure_Programming/Examples/API_Examples
16:58	blbrown	I was looking at stuart hollaway's blog and thinking, dang these are good examples...he should write a book on this stuff. Lo and behold he is
17:04	blbrown	http://paste.lisp.org/display/74655 is this not valid code
17:07	durka42	you might have missed some parens around the println
17:07	durka42	which line is 177?
17:07	gnuvince_	blbrown: use e#
17:07	ayrnieu	(let [e (gensym)] `(try ... (catch Exception ~e (. ~e ...))))
17:07	blbrown	durka42 error at when I called it (when-try (clojure.xml/parse db-path))
17:08	ayrnieu	if you really want 'e, you can ~'e
17:09	durka42	i mean there's no real need for a gensym there
17:09	ayrnieu	,[`(list ~'e) `(list e)]
17:09	clojurebot	[(clojure.core/list e) (clojure.core/list clojure.core/e)]
17:20	blbrown	this is what I ended up with, works. (defmacro when-try [body] `(try (~@body) (catch Exception ~'e nil)))
17:22	Chouser	sorry I'm jumping in late here, but why ~'e instead of e# ?
17:22	blbrown	and I guess I only need ~@body not (~@body)
17:23	ayrnieu	~body not (~@body)
17:23	clojurebot	It's greek to me.
17:25	blbrown	~your mom
17:25	clojurebot	Huh?
17:25	blbrown	~mom
17:25	clojurebot	It's greek to me.
17:26	durka42	Chouser: there's no user code being run with e bound, so why bother gensymming?
17:27	blbrown	I just pjb3 I just twittered you, small world isn't it
17:28	pjb3	blbrown: indeed
17:33	Chouser	durka42: hm, I guess I tend to think it the other way around. There's no need to give 'e' a particular name, so why bother fancy-quoting it?
17:33	gnuvince_	,(= Short java.lang.Short)
17:33	clojurebot	true
17:34	gnuvince_	hmmm
17:34	durka42	,(time (dotimes [_ 1000] (gensym)))
17:34	clojurebot	"Elapsed time: 12.013 msecs"
17:35	durka42	,(time (dotimes [_ 1000] `(~'e)))
17:35	clojurebot	"Elapsed time: 1.997 msecs"
17:36	ayrnieu	10ms difference at compile-time per thousand. I'll get right on optimizing for that.
17:36	gnuvince_	hahah :)
17:37	durka42	kind of a silly reason though
17:37	durka42	i guess if you refactored the macro later and bound e before user code, then forgetting to change it to e# could cause problems
17:37	durka42	:)
17:53	durka42	i must find a way to work "absquatulate" into normal conversation
17:57	Chouser	wouldn't it immediately cease to be a normal conversation?
17:57	durka42	quite possibly
17:57	Chouser	all the better, I suppose.
17:59	gnuvince_	durka42: what does that mean?
17:59	durka42	it means to leave abruptly
18:00	gnuvince_	Easy
18:00	gnuvince_	Go to a friend's
18:00	gnuvince_	Pretend you have an urgent phone call on your cell
18:00	gnuvince_	"I'm sorry, I must absquatulate!"
18:01	hiredman	"I cannot help but notice you are absquatulating with all the food"
18:08	hiredman	I forget, is there a pretty printer yet?
18:08	ayrnieu	there's cl-format
18:09	hiredman	does that do indenting and what not?
18:09	ayrnieu	I haven't explored it.
18:09	hiredman	where is there?
18:09	ayrnieu	github.
18:10	durka42	i believe "think about a pretty printer" is on the cl-format todo list
18:10	hiredman	damn
18:17	cooldude127	Woo, colloquy for iPhone :)
18:21	dreish	clojurebot: clojure-mode?
18:21	clojurebot	clojure-mode is an Emacs mode for Clojure, found at git://github.com/jochu/clojure-mode.git
18:21	dreish	Thanks.
18:22	durka42	to thank clojurebot,
18:22	durka42	clojurebot: botsnack
18:22	clojurebot	thanks; that was delicious. (nom nom nom)
18:36	eee	hi
18:37	eee	here's my latest gripe: I api is gibberish to me
18:37	eee	let's take a simple example that I understand
18:38	eee	hi
18:38	eee	about to complain :)
18:38	Chousuke	gibberish? you mean you can't understand it?
18:38	eee	yup
18:38	Chousuke	I guess it could be a bit better
18:38	eee	here's an example
18:38	Chousuke	You should read the wiki to get started; the api is more of a reference.
18:38	eee	look at "doall" . . .with doall, I know what it does
18:38	eee	and I understand the english description
18:38	eee	very nice
18:38	eee	so I'm lucky
18:39	eee	cause I can't make heads or tails of the gibberish part
18:39	Chousuke	which part is that?
18:39	eee	([coll] [n coll]) or something?
18:39	eee	what the heck?
18:39	Chousuke	ah, that's the possible parameter lists
18:39	ayrnieu	(defn foo ([a] (mangle a)) ([a b] (frob a b)))
18:40	Chousuke	eee: that means it can take either two parameters, or one
18:40	eee	oh
18:40	eee	should never use the brckets there
18:41	eee	shold be, like (<coll> <n coll>)
18:41	Chousuke	um, no
18:41	eee	brackets are actually used in the language
18:41	ayrnieu	eee, the 'foo' I just defined is valid clojure.
18:41	eee	yes I get it
18:41	Chousuke	eee: that's why the parameter lists are like that too
18:41	eee	it takes an a or an a and a b
18:41	Chousuke	they are like they are in the language itself.
18:42	eee	i've made functions like that
18:42	eee	I assume that's a multimethod, but I don't know
18:42	Chousuke	it's not a multimethod
18:42	eee	oh drat
18:42	Chousuke	it's simply an overloaded function
18:42	Chousuke	anyway, for complex parameters, using the language's own notation is better.
18:42	eee	yeah . . .I was hoping that was a multimethod
18:42	eee	oh well
18:42	dreish	eee: Have you read Programming Clojure?
18:43	eee	not yet
18:43	dreish	If you're interested enough to ask this many questions, why not buy the book?
18:43	eee	doesn't make sense to use something that's used in the language cause you might take it literally but I guess I understand now
18:43	dreish	It's an excellent way to learn the language.
18:43	Chousuke	for example, if you say your parameter list is a [[a b c] & rest] someone familiar with clojure and destructuring can easily tell how to call it.
18:44	eee	i didn't know that was destructureing. that's cool . . that fills in another blank. I use that form
18:44	eee	ok, so how would the docs look for something that destructures like that?
18:45	Chousuke	it would say its possible parameters are ([[a b c] % rest])
18:45	ayrnieu	but: destructuring doesn't work on lists, and it isn't coll-agnostic.
18:45	Chousuke	er
18:45	Chousuke	&*
18:46	Chousuke	eee: so you see, the parameter list in the doc exactly matches the function definition :)
18:46	eee	i've used [a b c & rest]
18:46	Chousuke	eee: that's a different function
18:46	eee	yeah
18:46	Chousuke	er, a different kind of parameter list that is.
18:46	Chousuke	my function, you'd call like (fun [1 2 3] 4 5 6)
18:47	eee	well, it doesn't exactly match the function definition because it's missing the context of the defn and the name of the function
18:47	Chousuke	so a, b and c would be 1, 2 and 3 respectively, and rest would be (4 5 6)
18:47	eee	better docs would be
18:47	Chousuke	eee: the name of the function is printed above isn't it.
18:47	eee	(defn foo [a] ... [a b] ...)
18:48	eee	now THAT would rock
18:48	eee	what I just wrote leaves no ambiguity
18:48	Chousuke	where would you put the documentation? :/
18:48	Chousuke	also I don't understand why this is at all better.
18:49	Chousuke	it's basically hte same.
18:49	Chousuke	just with a defn and a few parentheses.
18:49	eee	where it says "arguments" . . . .that's not quite accutate
18:49	durka42	it is ambiguous... is [a b] another param list, or is the vector part of the function
18:49	eee	there's a needless list around it . . .a set of parens confusing things
18:49	eee	night as well throw the defn in there
18:49	Chousuke	eee: it's not a needless list
18:49	Chousuke	it's a list of the possible argument vectors you can pass in
18:50	eee	well, I guess I get it now . . .but I would never have needed to my way. would have had only one interpretation
18:50	Chousuke	I don't see how :/
18:50	eee	what durka42 said is why
18:51	eee	it's awash with symbols . . . .
18:51	eee	anti-tufte
18:51	Chousuke	I don't follow :/
18:52	eee	that () around the whole thing
18:52	eee	I'd even go for
18:52	Chousuke	eee: well, that's because it's a list.
18:52	Chousuke	it's a list of the ways to call the function
18:52	eee	arguments: a \| a b
18:52	Chousuke	the list is more clojurey :/
18:52	eee	the vector isn't even needed
18:52	Chousuke	yes it is.
18:53	eee	just use a pipe to separate the two choices
18:53	eee	otherwise I'm thinking . . ."takes a vector with a collection, what?"
18:53	eee	noise
18:53	eee	a \| a b ...
18:53	eee	or
18:53	eee	in the case of doall
18:53	eee	coll \| coll1 coll2 ...
18:54	eee	coll \| coll1 coll2 ... colln
18:54	ayrnieu	so it takes a, a pipe, and ab . Or does it pipe a into a b? Is this like the unix shell? I'M CONFUSED.
18:54	Chousuke	eee: it's not like that though
18:54	eee	nowlol
18:54	Chousuke	it's coll \| n coll
18:54	eee	doall?
18:54	Chousuke	yeah
18:54	eee	i upset rhickey
18:54	Chousuke	the thing is, anyone familiar with how to define clojure function definitions can read the docs :/
18:54	eee	he's not gonna let me play
18:54	Chousuke	...
18:54	Chousuke	how to define clojure functions*
18:55	eee	ok, awesome . . .one more thing people have to learn
18:55	eee	I'm trying to help this stuff be how we all program
18:55	eee	cause I like it
18:55	eee	too bad rich missed that part
18:55	eee	:(
18:55	eee	in unix
18:55	Chousuke	well... clojure is not going to be very useful to you if you don't learn how to define clojure functions.
18:56	eee	angle brackets are reserved for <variable>
18:56	eee	so maybe
18:56	hiredman	it being functional and all
18:56	ayrnieu	eee, it's useful to distinguish between newbie problems (oh, now that I know X, I'll never worry about this ever again) and real, persistent problems. And, correspondingly, newbie-problem improvements and real improvements.
18:56	Chousuke	in C docs, the functions are just as often given as "void foo(bar, baz)" in docs, so I don't see your point.
18:56	Chousuke	clojure is not the unix shell.
18:56	eee	<coll> \| <coll1> <coll2> ... <colln>
18:57	eee	you should see my point
18:57	eee	in C
18:57	eee	you just see the prototype
18:57	eee	without the body
18:57	eee	but in clojure it's all split up
18:57	eee	the name
18:57	eee	then a different section with a list of params
18:57	Chousuke	they're not that far apart...
18:57	eee	why put the parens like it is a c function
18:57	ayrnieu	it's not just split up in the doc, it's split up in the function.
18:58	eee	when you write it, you use clojure syntax
18:58	Chousuke	eee: eh, the parens aren't ther for that purpose. it's a list, you see.
18:58	Chousuke	so you use parens.
18:58	eee	when you see it in the api, you see C syntax
18:58	eee	the parens wrap the params for some reason
18:58	eee	ok
18:58	eee	a list of optional cases
18:58	durka42	"some reason" being so that the compiler can parse it correctly
18:58	durka42	yes
18:59	gnuvince_	Did I miss something interesting/inflamatory?
18:59	eee	I dunno. just seems like if a prototype makes sense in c of java, clojure should do it like I wrote it
18:59	Chousuke	eee: that list of argument vectors actually exists as a data structure in the fuction metadata.
18:59	eee	(defn foo [a] ... [a b] ...)
18:59	Chousuke	the documentation just prints it as is.
18:59	eee	like that right there
19:00	eee	,(doc doall)
19:00	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 rests of the seq, retains the head and returns it, thus causing the entire seq to reside in memory at one time."
19:00	Chousuke	eee: but the problem is that that's not a prototype; that's a definition
19:00	hiredman	,(:arglists (meta #'println))
19:00	clojurebot	([& more])
19:01	eee	ok, if I ask for arglist, that's one thing
19:01	eee	if I ask for api, that's another
19:01	Chousuke	I really don't see a problem.
19:01	Chousuke	it's a convention.
19:01	Chousuke	get used to it.
19:01	eee	looks like all someone did for api was call arglist and comments
19:01	ayrnieu	so write a my-doc and show us how much better it is.
19:01	eee	convention . . . this is why I brought up tufte
19:01	Chousuke	eee: yeah, that's pretty much what it is.
19:02	hiredman	,(:arglists (meta #'my-doc))
19:02	clojurebot	java.lang.Exception: Unable to resolve var: my-doc in this context
19:02	hiredman	,(:arglists (meta #'mydoc))
19:02	clojurebot	java.lang.Exception: Unable to resolve var: mydoc in this context
19:02	hiredman	,(:arglists (meta #'doc))
19:02	clojurebot	([name])
19:02	hiredman	hmmm
19:02	Chousuke	eee: but what the arglist looks like is the least of the documentation worries.
19:02	eee	,(arglist doall)
19:02	clojurebot	java.lang.Exception: Unable to resolve symbol: arglist in this context
19:02	eee	,(arglist #doall)
19:02	clojurebot	Eval-in-box threw an exception:No dispatch macro for: d
19:02	eee	,(:arglist #doall)
19:02	clojurebot	Eval-in-box threw an exception:No dispatch macro for: d
19:02	Chousuke	,(:arglists ^#'doc)
19:02	clojurebot	([name])
19:03	Chousuke	,(:arglists ^#'doall) ;een
19:03	clojurebot	([coll] [n coll])
19:03	Chousuke	even :P
19:03	eee	,(:arglist ^#'doall)
19:03	clojurebot	nil
19:03	Chousuke	missed an s
19:03	eee	,(:arglists ^#'doall)
19:03	clojurebot	([coll] [n coll])
19:03	eee	oh
19:04	eee	thanks for explaining it to me
19:04	Chousuke	you're free to go write your own pretty-printer for documentation.
19:04	Chousuke	something searchable or cross-referenced would be great. the regular API page is a bit plain :/
19:04	lisppaste8	ayrnieu pasted "print-doc" at http://paste.lisp.org/display/74659
19:04	eee	sounds like a good thing to practice with
19:05	gnuvince_	Chousuke: pretty sure there was someone who did something like that in Enclojure and posted about it on the mailing list.
19:05	eee	i know several people added doc writers to that list in the user group about third-party stuff
19:05	eee	well, I was just complaining about the enclojure stuff
19:05	eee	that's what brought me in here
19:05	eee	:)
19:06	eee	all I know . . .is that I didn't feel like I had to ask this question for java. We can call it a broken language if we like
19:06	eee	but for some reason
19:06	eee	people are able to get up and going on it
19:06	eee	without much understanding of "convention"
19:07	eee	but I get it now
19:07	eee	(I think)
19:07	eee	:)
19:07	eee	,(doc arglists)
19:07	clojurebot	java.lang.Exception: Unable to resolve var: arglists in this context
19:08	eee	oh
19:08	eee	taht's a key
19:08	eee	huh?
19:08	ayrnieu	,(:a {:a :b})
19:08	clojurebot	:b
19:08	gnuvince_	It's a key on the meta-data.
19:08	eee	yeah
19:08	Chousuke	,^#'doall
19:08	clojurebot	{:ns #<Namespace clojure.core>, :name doall, :file "core.clj", :line 1675, :arglists ([coll] [n coll]), :doc "When lazy sequences are produced via functions that have side\n effects, any effects other than those needed to produce the first\n element in the seq do not occur until the seq is consumed. doall can\n be used to force any effects. Walks through the successive rests of\n the seq, retains the head and returns
19:09	eee	now can functions be named with colon at the beginning?
19:09	Chousuke	no
19:09	gnuvince_	But what was your objection? That Clojure doesn't have a Java-like syntax?
19:09	ayrnieu	,(let [x :a] [(class x) (x {:a :b})])
19:09	clojurebot	[clojure.lang.Keyword :b]
19:09	eee	no
19:09	eee	that the api makes sense
19:09	eee	in java
19:09	eee	without any confusing
19:09	Chousuke	symbols starting with a colon are always keywords :)
19:09	eee	now I have to do a survey of newbies
19:10	gnuvince_	That the API makes sense?
19:10	eee	and find out if it's just that I'm the only one who finally asked out loud
19:10	gnuvince_	The API or the API documentation?
19:10	Chousuke	maybe some note could be put on top of the API documentation page about how to read it
19:10	ayrnieu	,[(class (symbol ":a")) (class :a)]
19:10	clojurebot	[clojure.lang.Symbol clojure.lang.Keyword]
19:10	eee	i mean the format of the documentation
19:10	Chousuke	it's easy to read, but you do need to know how
19:10	ayrnieu	,(= (symbol ":a") :a)
19:10	clojurebot	false
19:10	Chousuke	or hm, actually
19:10	eee	that would be good chosuke . . .I actually looked for that
19:10	gnuvince_	eee: on the web site? Yeah, it's not terrible. Feel free to contribute something better organized/indexed.
19:10	eee	maybe it is there somewhere
19:11	Chousuke	the API page itself doesn't have your particular problem
19:11	Chousuke	but (doc foo) does :/
19:11	gnuvince_	Chousuke: how do?
19:11	gnuvince_	*so
19:11	Chousuke	gnuvince_: he doesn't like the way argument vectors are just dumped out as a list
19:11	eee	\:
19:12	eee	i wish I cold be so succinct. well put
19:12	Chousuke	the api page actually has (doall coll), (doall n coll)
19:12	eee	heh
19:12	gnuvince_	ah
19:12	eee	i don't even remember . . .I guess it's how it is in enclojure them
19:12	ayrnieu	eee, did you see that paste? http://paste.lisp.org/display/74659 is all the heavy lifting that doc does wrt. documentation. So if you want something better, implement it and see if it matters.
19:13	eee	i just reflected on how I didn't get the api page, either . . .so gave up on it along time ago
19:13	gnuvince_	I figure once you learn it, the (doc) format makes sense; a list of all the different arities a function takes.
19:14	Chousuke	yeah.
19:15	eee	you know how people have legends in the corner of their slides sometimes in presentations? this guy http://www.edwardtufte.com/tufte/ has a lot to say about making stuff self-defining.
19:15	eee	no conventions
19:15	eee	cognitively ergonomic
19:16	gnuvince_	Write a patch
19:16	eee	well, I'm see what people think
19:16	eee	I could just be an idiot
19:16	eee	but it could be fun to play with
19:16	eee	I grabbed that code
19:17	gnuvince_	like ayrnieu said, just go and edit print-doc
19:19	eee	k, thanks
19:19	ayrnieu	,[:a ::a]
19:19	clojurebot	[:a :clojure.core/a]
19:20	eee	is print-doc part of doc?
19:20	eee	,(doc print-doc)
19:20	clojurebot	"([v]); "
19:20	Chousuke	not documented :)
19:21	gnuvince_	http://code.google.com/p/clojure/source/browse/trunk/src/clj/clojure/core.clj#2766
19:21	gnuvince_	print-doc is called by the doc macro
19:22	Chousuke	used by doc; if you do (binding [print-doc my-print-doc] (doc foo)) you might be able to override it
19:22	Chousuke	straight up redefining it works too, but it's less clean :/
19:22	gnuvince_	,'#{a b c}
19:22	clojurebot	#{a c b}
19:23	eee	it barfed when I tried to redefine it
19:23	hiredman	you can actually rebind the var the doc macro is attached to too
19:23	eee	ok
19:23	hiredman	(binding [doc nil] (doc foo))
19:23	ayrnieu	,(binding [print-doc #(println (:name ^%) " - " (:arglists ^%))] (doc doall))
19:23	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 rests of the seq, retains the head and returns it, thus causing the entire seq to reside in memory at one time."
19:24	ayrnieu	WFM. clojurebot is doing something special.
19:24	hiredman	clojurebot rebinds doc internally
19:24	hiredman	to another macro
19:25	eee	that bot is cool. who wrote it?
19:25	ayrnieu	,(binding [print-doc #(println (:name ^%) " - " (:arglists ^%))] (clojure.core/doc doall))
19:25	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 rests of the seq, retains the head and returns it, thus causing the entire seq to reside in memory at one time."
19:25	ayrnieu	oh well.
19:25	Chousuke	hiredman wrote most of it. I did some restructuring and now hiredman seems to be implementing all kinds of weird features.
19:26	hiredman	ayrnieu: it rebinds doc which in that namespace refers to clojure.core/doc, so in that dynamic scope any reference to clojure.core/doc or doc is something else
19:31	gnuvince_	When interfacing with Java methods that might raise exceptions, what's the "better" way do deal with that in Clojure? Let them be thrown and let user code deal with it or deal with it at the library level and return nil?
19:32	gnuvince_	More concretely, I am dealing with java.nio.ByteBuffers, and I was wondering what would be the right way to deal with BufferUnderflowExceptions.
19:32	hiredman	,(pipeline '(a b (h (g y))))
19:32	clojurebot	java.lang.IllegalArgumentException: Don't know how to create ISeq from: Symbol
19:33	hiredman	,(pipeline '(a b (h y (g y))))
19:33	clojurebot	java.lang.IllegalArgumentException: Don't know how to create ISeq from: Symbol
19:33	hiredman	hmmm
19:33	ayrnieu	,(doc pipeline)
19:33	clojurebot	"([a]); "
19:35	blbrown	is 'doto' a macro?
19:35	Chousuke	yes
19:35	blbrown	I like it
19:35	Chousuke	see -> as well :)
19:36	Chousuke	gnuvince_: I wonder; does returning nil in such a case make sense?
19:36	Chousuke	gnuvince_: are there cases where returning will could be useful non-error behaviour?
19:37	Chousuke	... nil
19:37	ayrnieu	,(symbol (.getName (first (.getMethods String))))
19:37	clojurebot	hashCode
19:37	ayrnieu	,(-> String .getMethods first .getName symbol)
19:37	clojurebot	hashCode
19:39	gnuvince_	Chousuke: not really. Actually, now that I think about it, I think that an exception indicates that there's something very wrong
19:41	Chousuke	yeah.
19:41	ayrnieu	extraordinary control flow
19:42	Chousuke	exceptions should indicate that not everything goes as planned. returning nil instead is a "I have nothing for you" kind of thing
19:42	gnuvince_	Yeah
19:43	gnuvince_	I guess that if you're expecting to read 10 bytes of data from a buffer and there are only 8, it should be an exception.
19:44	Chousuke	only if the missing two bytes went missing because of a malfunction :/
19:48	gnuvince_	Chousuke: it's from a defined file format. If there's an error anywhere, the entire thing is corrupt.
19:48	Chousuke	yeah, that sounds exceptional enough :)
19:53	gnuvince_	Chousuke: I don't know if you read my latest blog post about reading binary data, but it's the continuation of that.
19:53	gnuvince_	Changed a few things
19:53	Chousuke	gnuvince_: maybe it'd still make sense to return nil in the case the file is corrupt
19:54	gnuvince_	Chousuke: I will probably return nil for the whole operation
19:54	gnuvince_	(decoding the file)
19:55	gnuvince_	But not for the smallest tasks like reading one field.
19:55	gnuvince_	But at that point, I'll be in consumer code.
19:55	gnuvince_	Where making that decision is easier.
20:05	hiredman	clojurebot: translate to it: my hovercraft is full of eels!
20:05	clojurebot	il mio hovercraft è pieno di anguille!
20:07	hiredman	clojurebot: translate to zh: my hovercraft is full of eels!
20:07	clojurebot	我国气垫船充满鳗鱼！
20:14	blbrown	has anyone built a web application with clojure and spring? I wonder if it is just better to avoid all of those java web frameworks all together? hmm
20:32	danlarkin	clojurebot: translate to hu: my hovercraft is full of eels!
20:32	clojurebot	Saját légpárnás tele van eles!
20:32	danlarkin	not handling the encoding very well I guess
20:36	ivan	double-utf8ed
21:14	hiredman	~translate to ru: in soviet russia the state mutates you!
21:14	clojurebot	? ????????? ?????? ??????????????? ???????? ???!
21:14	hiredman	I guess puns don't translate well then...
21:14	ayrnieu	~translate to zh: ? ????????? ?????? ??????????????? ???????? ???!
21:14	clojurebot	? ????????? ?????? ??????????????? ???????? ??? !
21:14	hiredman	ayrnieu: to/from english only
21:15	ayrnieu	~translate from ru: ? ????????? ?????? ??????????????? ???????? ???!
21:15	clojurebot	In Soviet Russia the state mutates you!
21:15	danlarkin	clojurebot: translate to hu: my hovercraft is full of eels!
21:15	clojurebot	Saj�t sikl�haj� tele van eles!
21:15	danlarkin	heyy there were are
21:16	ayrnieu	~translate from zh: ??????
21:16	clojurebot	What you are
21:16	durka42	a friend who speaks russian tells me that google is spouting nonsense
21:16	ayrnieu	(not quite)
21:22	Wizardofwestmarc	a translation program spewing gibberish? Inconceivable!
21:23	blbrown	clojure is awesome. Are the developers working with any companies to integrate the technology. I think that is where a lot of open source languages/software lose out. They don't consider businesses.
21:23	blbrown	I will probably say something about clojure in a developer meeting, but that wont be for months
21:25	durka42	there was a post about clojure being used in a hospital
21:26	Wizardofwestmarc	well, not probably quite what you mean but Rich did present at a recent JVM workshop hosted by Sun themselves
21:26	blbrown	durka42, I remember that
21:26	Wizardofwestmarc	That sorta thing helps get the business side rolling
21:26	blbrown	I saw part of the google tech talk
21:26	blbrown	...or was that another language, I am getting them all confused these days
21:27	Wizardofwestmarc	hm, don't recall a GTT
21:28	blbrown	maybe not
21:30	Wizardofwestmarc	although Google taking an interest in clojure would certainly be interesting if it has happened/were to happen
21:30	Wizardofwestmarc	since they already have Java in their standard dev stack
21:31	blbrown	they seem to have low standards on who they invite though (IMHO)
21:31	Wizardofwestmarc	oh? I've only watched a few of the talks of theirs
21:31	blbrown	OK, president Obama doesn't count
21:31	Wizardofwestmarc	(GTTs)
21:31	Wizardofwestmarc	the one for the xkcd cartoonist was interesting because a ton of big name guys in comp sci/programming showed up
21:32	blbrown	I mean they have a high quantity over quality ratio
21:32	Wizardofwestmarc	right
21:32	Wizardofwestmarc	I usually just watch ones by people I figure are worth watching based on their reputation
21:32	blbrown	I think the JVM workshop is probably a bigger deal
21:33	ayrnieu	"In this chapter you have seen four concurrency APIs: refs, atoms, agents,
21:33	ayrnieu	blech.
21:33	ayrnieu	" and vars. [...] So we are left with good old-fashioned locking."
21:33	Wizardofwestmarc	for getting noticed by a wide audience of companies who use the java stack it probably is
21:33	Wizardofwestmarc	ayrnieu: Halloway's book?
21:33	ayrnieu	Yes.
21:34	Wizardofwestmarc	heh
21:34	Wizardofwestmarc	I still need to read the newest beta
21:34	Wizardofwestmarc	skimmed the chapter on Functional, but haven't looked at the other changes
21:34	Wizardofwestmarc	also, was rather sad to see the Monte Carlo chapter got cut
21:35	blbrown	Wizardofwestmarc, there are lot of missing chapters
21:35	Wizardofwestmarc	beta 6 I think is content complete
21:35	blbrown	hell, I could have been happy with just that chapter
21:35	blbrown	oooo, nice
21:36	Wizardofwestmarc	the impression I get is it's gone >(
21:36	blbrown	that sucks
21:36	Wizardofwestmarc	but yes, 5 and before had a bunch of stuff missing still
21:36	blbrown	they never put anything production quality in a tech book
21:36	Wizardofwestmarc	true
21:38	Wizardofwestmarc	but boy, do I keep buying them <_<
21:38	Wizardofwestmarc	was debating grabbing o'reily's new book, Algorithms in a nutshell
21:39	blbrown	Wizardofwestmarc, me too, I finally got code complete 2
21:39	Wizardofwestmarc	GREAT book
21:40	blbrown	Wizardofwestmarc, that book looks interesting
21:40	Wizardofwestmarc	I reread it every couple years
21:40	Wizardofwestmarc	CC2 and Pragmatic Programmers are both books I dig out of my bookshelf every so often as refreshers
21:40	blbrown	I think I may have an audit file with the list of checklists at the end of each chapter "DId you refactor this code? yes/no" And I am guessing it could apply to clojure as well
23:01	dreish	Random fireworks outside.
23:44	blbrown	Wizardofwestmarc, I got the algorithm book.
23:45	blbrown	actually I am thinking of writing a book (JVM languages for Enterprise Development, will cover a small bit in clojure)
23:50	gnuvince_	Is it possible that elements in a Seq and in a vector are in opposite orders? If I don't call (into [] data) for one of my functions, my other functions yield erroneous results.
23:53	durka42	what are the other functions?
23:53	durka42	some functions treat lists and vectors as having opposite order
23:53	durka42	,(pop [1 2 3])
23:53	clojurebot	[1 2]
23:54	durka42	,(pop '(1 2 3))
23:54	clojurebot	(2 3)
23:54	durka42	,(conj [1 2 3] 4)
23:54	clojurebot	[1 2 3 4]
23:54	durka42	,(conj '(1 2 3) 4)
23:54	clojurebot	(4 1 2 3)
23:54	gnuvince_	durka42: some functions inside my program, not Clojure ones.
23:55	gnuvince_	I'm trying to find out if I can use Seqs, because converting into vectors takes quite a lot of time: to process 1000 files, seqs takes 1.6 second (with the wrong results though) and vectors take 16 seconds
23:55	durka42	well, i meant if your functions depend on clojure.core fns such as the above
23:55	gnuvince_	So obviously, I'd want to have my cake and eat it too.
23:55	durka42	want to paste something?
23:56	gnuvince_	durka42: It's quite a lot of code.
23:56	gnuvince_	I'm going to try and narrow it down
23:56	gnuvince_	At worse, I'll put it on github (where it was going anyway)
23:59	gnuvince_	HAH!
23:59	gnuvince_	Got it
23:59	durka42	got it?
23:59	gnuvince_	doall was needed