#clojure – 2015-04-18

2015-04-18

00:59	lexi-lambda	Anyone up for a simple code review? ;D http://codereview.stackexchange.com/q/87277/12229
01:12	seancorfield	lexi-lambda: I'm curious about the maybe fn -- seems to only be used with string? and (string? nil) is false so that seems safe without (maybe string?)
01:13	lexi-lambda	seancorfield: But ((maybe string?) nil) is true.
01:13	seancorfield	oh, i see
01:14	seancorfield	interesting... i misunderstood the intent...
01:14	seancorfield	the only other thing would be making the functions public instead of private -- they might be useful to users of the library
01:15	seancorfield	the community consensus seems to be to prefer public over private unless there's a really good argument to keep something private
01:15	seancorfield	looks pretty clean to me apart from that
01:15	lexi-lambda	Okay, good to know. I think you're right that query-string->map and those sorts of functions should probably be public.
01:15	lexi-lambda	What about `maybe`? I don't think making that public would make much sense.
01:16	lexi-lambda	It would make sense to put it in a separate utils lib or something, but I didn't do that for that one function.
01:18	seancorfield	perhaps create a specific fn for the intended "maybe string?" where nil is true?
01:18	clojurebot	Pardon?
01:19	seancorfield	(unless you use maybe with something other than string?)
01:20	lexi-lambda	seancorfield: I used to, I think, but it ended up getting unused. I like the way (maybe string?) expresses the "string-or-nil" concept, though.
01:21	seancorfield	how about (defn may-be-string? [s] (or (nil? s) (string? s)))
01:21	seancorfield	hard to pick a good name for that
01:22	seancorfield	optional-string?
01:30	lexi-lambda	"maybe" is definitely the Haskell-y way to phrase it, so "optional" might be a better name
01:34	seancorfield	lexi-lambda: naming is hard :)
03:50	kaiyin	does core.typed boost performance?
03:56	atomi	anyone know a good blog post for a getting started with http-kit?
03:56	atomi	I'm new to clojure all together and I just did the lein new hello
03:57	atomi	I got -main going and did lein uberjar
03:57	atomi	worked just fine :)
04:01	atomi	the http-kit examples aren't complete from a new user context
04:04	amalloy	kaiyin: it's just a static analysis tool. it doesn't make any difference to the compiled code
04:05	atomi	I added http-kit "2.1.19" to :dependencies in my project.clj
04:08	atomi	okay how about a good getting started book for clojure
04:12	atomi	haha nevermind I got some good one
04:12	atomi	s
04:57	lbradstreet	Is it possible to use with-redefs for defmethods?
05:13	darthdeus	is there an ordering guarantee on channels? such as if 3 producers are blocking to put a value in, and I take one out, do I take out the first one?
05:28	amalloy	darthdeus: i don't see how you could even observe whether that is actually happening or not, from within the channel ecosystem. how do you know which one is first?
05:28	darthdeus	amalloy: I'm using channels to synchronize some stuff in my tests, and I need to have some ordering guarantees
05:29	darthdeus	so I'm asking if core.async already does this, or if I need to roll my own
05:31	amalloy	darthdeus: what i'm getting at is, there's no way oyu can tell whether ordering is preserved or not. try to devise an experiment which will behave one way if channels are ordered in the way you describe, and behave differently if they're not
05:31	darthdeus	amalloy: well I've done that, but I'm not sure if I'm just getting lucky, or if this is by design
05:31	darthdeus	and I can't find it anywhere in the docs
05:31	amalloy	what is your experiment?
05:33	darthdeus	amalloy: http://stackoverflow.com/questions/29715462/if-multiple-subscribers-block-on-a-core-async-channel-is-the-order-in-which-the
05:35	amalloy	mmm. well yes, you can observe the difference if you're using the super low-level stuff like put! and take!. but you're not supposed to, right, in general? like you can't write this test with <! and >!
05:36	amalloy	anyway i don't know the answer to the question you're asking, but i'm encouraging you to think aobut whether it is really a question that's good to ask
05:38	darthdeus	amalloy: well, what I'm really doing is a writing a test, where multiple websocket clients (which are in clojure) talk to a server, and then I need to assert on the messages they get
05:39	darthdeus	and I'm using channels to gather incoming responses for each client, and then assert on them
05:40	darthdeus	I'm not really sure how to do this in any other way
05:40	darthdeus	I mean, I need the clients to run both at the same time, and individually collect incoming events in the order in which they come
05:42	darthdeus	I guess I could do the asserts in a (go) block, hmm
05:42	darthdeus	but that wouldn't really change much
05:42	darthdeus	since I need to synchronize across multiple async clients
06:29	H4ns	can anyone recommend a very simple, serverless, file backed persistence library for clojure?
06:31	expez	H4ns: spit, slurp and edn/read-string
06:31	H4ns	expez: right. i can roll my own, but i'd prefer not to.
06:33	expez	What functionality do you need which isn't provided by that solution?
06:33	H4ns	expez: serialization that works even when i have print-length set, for example
06:34	H4ns	expez: there are several fine points that one needs to get right.
06:36	H4ns	expez: also, when you're serializing all data into one file, you'll probably want to make sure that your writes succeed, and maybe keep a few previous versions etc. etc.
06:38	expez	H4ns: https://github.com/alandipert/enduro this might fit your bill. I haven't tried it out myself but it's in my "hammer looking for nail" pile :)
06:39	H4ns	expez: ah, that looks neat, thanks!
06:51	H4ns	expez: enduro is precisely what i needed, thank you again!
07:10	_1_	hyy anyone from patnaa
08:46	kaiyin	how do you exit a running function in lein repl?
08:47	TEttinger	kaiyin, ctrl-d on windows exits the whole repl
08:47	TEttinger	not sure if there's a way to only kill the current fn
08:47	TEttinger	(there might be?)
08:47	TEttinger	not sure if ctrl-d also works on linux/mac
08:48	TEttinger	(I guess it probably does since ctrl-c is force kill everywhere)
08:52	kaiyin	doesn't work, neither c-d nor c-c.
08:52	darthdeus	hey guys, any thoughts? https://github.com/technomancy/leiningen/issues/1884
08:53	TEttinger	kaiyin, try c-c c-c
08:53	kaiyin	TEttinger: tried that, too. not working.
08:53	TEttinger	that's uh not good. linux?
08:54	kaiyin	osx.
08:54	TEttinger	oh were you pressing cmd?
08:54	TEttinger	ctrl-c is the actual key on osx
08:54	kaiyin	no, ctrl.
08:54	TEttinger	ok
08:55	TEttinger	I am not sure then, there's probably some osx users who can answer better than I can
08:55	kaiyin	ok, thanks. :)
09:00	kaiyin	anyways, i had to shut down the terminal to exit that function. Is there a function out there for exiting its enclosing function when certain conditions are met?
09:00	kaiyin	I tried (System/exit 0), but it also shuts down the repl.
09:10	TEttinger	kaiyin, like break in loops in C-like languages, or an early return statement?
09:10	TEttinger	the closest equivalent to break for reduce calls is reduced, which is generally handy for lots of stuff
09:10	TEttinger	(doc reduced)
09:11	clojurebot	"([x]); Wraps x in a way such that a reduce will terminate with the value x"
09:11	kaiyin	TEttinger: yeah, like an early return statement.
09:12	TEttinger	I mean ideally clojure code is structured into lots of small fns, but that isn't always practical...
09:13	TEttinger	do you know what wasn't terminating?
09:13	TEttinger	like an evaluation of an infinite lazyseq?
09:14	TEttinger	if you specifically want to denote that there is an error state in your program, you can just throw an exception
09:14	TEttinger	(doc throw)
09:14	clojurebot	Gabh mo leithscéal?
09:14	TEttinger	I think it's a built-in
09:15	TEttinger	(throw (Exception. "Some text"))
09:17	kaiyin	TEttinger: ah, indeed, that's a nice idea.
09:17	TEttinger	an advantage of being built with Java, kinda :)
09:17	TEttinger	plenty of Exception classes to choose from, jhaha
09:18	H4ns	we're using slingshot instead of raw java exceptions: https://github.com/scgilardi/slingshot
09:19	TEttinger	indeed. there are lots of solutions out there, I've heard good things about slingshot
09:19	TEttinger	how do you like it, H4ns?
09:20	H4ns	TEttinger: much better than raw exceptions because in those cases where we wanted to catch some specific error, we did not need to subclass Exception. so it does what we need so far.
09:20	clojurebot	Titim gan éirí ort.
09:20	TEttinger	great!
09:20	TEttinger	(inc H4ns)
09:20	lazybot	⇒ 3
09:20	TEttinger	(inc scgilardi)
09:20	lazybot	⇒ 1
09:54	justin_smith	TEttinger: you can't do anything involving throw with clojurebot, not even ask for docs
09:54	justin_smith	security
09:55	TEttinger	wow
09:55	justin_smith	H4ns: it's not super hard to directly use ex-info
09:56	H4ns	justin_smith: yes. it is not super hard to use slingshot either. why should i not?
09:56	justin_smith	H4ns: I only mention this because dep creep
09:56	justin_smith	but if you are OK using slingshot that's fine, of course
09:56	TEttinger	,((fn rec [] (if (= (rand-int 100) 0) (throw (Exception. "wowza bo bowza!")) (rec))))
09:56	clojurebot	#error{:cause "wowza bo bowza!", :via [{:type java.lang.Exception, :message "wowza bo bowza!", :at [sandbox$eval25$rec__26 invoke "NO_SOURCE_FILE" 0]}], :trace [[sandbox$eval25$rec__26 invoke "NO_SOURCE_FILE" 0] [sandbox$eval25$rec__26 invoke "NO_SOURCE_FILE" 0] [sandbox$eval25$rec__26 invoke "NO_SOURCE_FILE" 0] [sandbox$eval25$rec__26 invoke "NO_SOURCE_FILE" 0] [sandbox$eval25$rec__26 invoke "NO_...
09:57	justin_smith	TEttinger: hrm, so that's not the thing I thought it was...
09:57	mavbozo	,(throw (ex-info "wowza bo bowza!" {}))
09:57	clojurebot	#error{:cause "wowza bo bowza!", :via [{:type clojure.lang.ExceptionInfo, :message "wowza bo bowza!", :at [clojure.core$ex_info invoke "core.clj" 4591]}], :trace [[clojure.core$ex_info invoke "core.clj" 4591] [sandbox$eval51 invoke "NO_SOURCE_FILE" 0] [clojure.lang.Compiler eval "Compiler.java" 6792] [clojure.lang.Compiler eval "Compiler.java" 6755] [clojure.core$eval invoke "core.clj" 3079] ...]}
09:57	TEttinger	&((fn rec [] (if (= (rand-int 100) 0) (throw (Exception. "wowza bo bowza!")) (rec))))
09:57	lazybot	java.lang.Exception: wowza bo bowza!
09:57	TEttinger	heh
09:58	mavbozo	&(ex-info "wowza bo bowza!" {})
09:58	lazybot	⇒ #<ExceptionInfo clojure.lang.ExceptionInfo: wowza bo bowza! {}>
10:02	mavbozo	&(throw (ex-info "wowza bo bowza!" {}))
10:02	lazybot	clojure.lang.ExceptionInfo: wowza bo bowza! {}
10:03	mavbozo	what's the difference between & and ,
10:25	kaiyin	why does this tic tac toe program fail to print the board? https://gist.github.com/kindlychung/bd9e46eda02784cec68d
10:27	xeqi	kaiyin: L42. map is lazy, so won't realize any of the printing unless the returned nil is needed
10:27	xeqi	(doc doseq)
10:27	clojurebot	"([seq-exprs & body]); Repeatedly executes body (presumably for side-effects) with bindings and filtering as provided by \"for\". Does not retain the head of the sequence. Returns nil."
10:27	xeqi	try doseq instead
10:28	kaiyin	xeqi: ok, i'll try that.
10:31	kaiyin	xeqi: cool, now it works.
10:32	TEttinger	mavbozo: & executes the next form with lazybot, while ',' uses clojurebot. clojurebot is running almost the latest 1.7 master, lazybot isn't as recent but doesn't restrict printlength
10:32	mavbozo	,clojure-version
10:32	clojurebot	{:major 1, :minor 7, :incremental 0, :qualifier "master", :interim true}
10:32	mavbozo	&clojure-version
10:32	lazybot	⇒ {:major 1, :minor 7, :incremental 0, :qualifier "alpha1"}
10:33	TEttinger	you can also use lazybot mid-message ##(take 1000 (range)) like that
10:33	lazybot	⇒ (0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 ... https://www.refheap.com/99759
10:33	mavbozo	TEttinger, ah, so that's the difference
10:33	mavbozo	w00t!
10:33	TEttinger	the refheap pasting is extremely handy
10:33	TEttinger	but clojurebot's restricted printlength lets it deal with infinite lazyseqs without dying
10:34	TEttinger	(by forgetting the "infinite" part :D)
10:37	mavbozo	really awesome
10:40	mavbozo	##(take 1000 (cycle [0 1 2]))
10:40	lazybot	⇒ (0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 ... https://www.refheap.com/99760
11:02	si14	is there any better way nowadays to hint object array type other than using string representation?
11:03	si14	into-array somehow kinda works sometimes, but I can't understand why and why it fails sometimes
11:04	si14	https://gist.github.com/si14/c5d584bf622e6de91348 here is an example — without that ugly array hint it warns me about reflection on that .set
11:09	Guest67	I'm trying to get transducers to work for first time, but when I call (map somefn) i get an arity exception. i'm using clojure 1.6. Any hint what i am missing?
11:09	mavbozo	Guest67, use clojure 1.7
11:10	mavbozo	,clojure-version
11:10	clojurebot	{:major 1, :minor 7, :incremental 0, :qualifier "master", :interim true}
11:10	mavbozo	,(map somefn)
11:10	clojurebot	#error{:cause "Unable to resolve symbol: somefn in this context", :via [{:type clojure.lang.Compiler$CompilerException, :message "java.lang.RuntimeException: Unable to resolve symbol: somefn in this context, compiling:(NO_SOURCE_PATH:0:0)", :at [clojure.lang.Compiler analyze "Compiler.java" 6543]} {:type java.lang.RuntimeException, :message "Unable to resolve symbol: somefn in this context", :at [...
11:10	mavbozo	,(map +)
11:10	clojurebot	#object[clojure.core$map$fn__4528 0x1dcd1bd6 "clojure.core$map$fn__4528@1dcd1bd6"]
11:10	mavbozo	&(map +)
11:10	lazybot	⇒ #<core$map$fn__4338 clojure.core$map$fn__4338@5442634c>
11:11	Guest67	thanks
12:02	nooga	Can I use locking with values like strings, eg. (locking "someid" ...) ?
12:10	greoih	Hi! I'm sorry if this is a newbie question. I'm trying to get started with Clojure & Clojurescript. Making a webapp with the lein template reagent-template (https://github.com/reagent-project/reagent-template). My problem is with Emacs Cider integration. Running `lein test` works fine, but when I from core.cljs (with cider jacked in) hit `C-c ,`, I get the error `No tests namespace: list-paddan.core-test`, even if that namespace does
12:10	greoih	exist.
12:10	greoih	So I wonder, how can you tell cider where to look for tests namespaces, or how does that even work? Thank you very much.
12:14	sdegutis	Are transducers just a new name for the common map-reduce pattern?
12:27	kaiyin	why does the snake stand still? https://gist.github.com/kindlychung/93b50754d75b8056b7eb
12:33	noncom\|2	sdegutis: no, i is a more general thing
12:34	J_Arcane	... Hmm. How do you push something to the top of an atom? doing swap! with cons takes the arguments in the wrong order, and I get a mess.
12:36	noncom\|2	J_Arcane: refs are for this
12:36	noncom\|2	use a ref + dosync i guess
12:37	noncom\|2	J_Arcane: however, i did not get into such a situation with atoms
12:37	noncom\|2	consing was always in the right order if i had just one thread to do this
12:37	noncom\|2	kaiyin: why don't you use seesaw?
12:38	kaiyin	noncom\|2: I am following some tutorial on the web, they didn't use seesaw.
12:38	kaiyin	I also think seesaw is a better option.
12:39	kaiyin	but that's something else to learn.
12:40	J_Arcane	noncom\|2: What I found is that if I've got a vector of strings in an atom, doing (swap! atm cons "string") actually tries to cons the contents of the atom onto "string", rather than the other way around in ClojureScript.
12:43	nuwanda__	J_Arcane: that's because:
12:43	nuwanda__	,(doc cons)
12:43	clojurebot	"([x seq]); Returns a new seq where x is the first element and seq is the rest."
12:43	noncom\|2	kaiyin: where is the update loop for your game?
12:44	nuwanda__	J_Arcane: you can do (swap! atom #(cons item %))
12:46	kaiyin	noncom\|2: i am not really sure, it's probably not there. Is that the problem?
12:46	kaiyin	I don't understand the gui part very well.
12:46	J_Arcane	nuwanda__: Ah! Right. I thought maybe that was the case with the argument order, just couldn't figure out a way around it.
12:46	noncom\|2	kaiyin: i cannot see where should it happen. i guess that the Timer is to be used to schedule updates on a regular basis...
12:47	kaiyin	noncom\|2: yeah, it's supposed to update the game state every 100 ms.
12:49	J_Arcane	IS there a 'flip' function in Clojure?
12:50	tomjack	there isn't a defined function for that in core, no
12:50	tomjack	I would call this _a_ flip function: (fn [f] (fn [x y] (f y x)))
12:51	tomjack	but then you might ask "what if f doesn't take two args?"
12:51	tomjack	and you will maybe understand why there is no defined 'flip' function in core :)
12:51	noncom\|2	kaiyin: well, it is not employed anywhere
12:52	noncom\|2	probably a piece of the program is missing
12:52	kaiyin	ok
12:52	tomjack	er, not "what if f doesn't take two args?" really, but "what if f also takes (not= two) args?"
12:56	noncom\|2	tomjack: the word "flip" implies duality
12:57	noncom\|2	tomjack: so, the case for more args would be like "scroll-args" or "reverse-args"
12:57	tomjack	or other things
13:03	tomjack	I was thinking 'flip' is less good a fit in an uncurried world, but I'm not sure that's really true
13:03	noncom\|2	tomjack: yeah, other things too
13:19	kaiyin	why does used-id contain duplicated entries here: https://gist.github.com/kindlychung/8770cc28cc76250071cd ?
13:23	lalaland1125	kaiyin, Don't you need to use an atom for uid?
13:25	lalaland1125	NVM, didn't see the ref.
13:25	kaiyin	lalaland1125: why?
13:25	kaiyin	ok
13:28	kaiyin	Since Refs are for Coordinated Synchronous, I don't understand why used-id should contain duplicated numbers.
13:29	tomjack	is commute the only thing you've tried?
13:31	tomjack	I think the answer to your question is "that's how commute works". if you are just learning refs, use alter or ref-set instead
13:31	tomjack	if you are learning commute, then I guess maybe you just did :)
13:36	kaiyin	how does commute work then?
13:36	postpunkjustin	I have some acceptance tests that I don't necessarily want to run all the time. They're marked with ^:acceptance, but "lein test" still runs them by default. Is there some way I can skip those by default?
13:36	kaiyin	tomjack: unfortunately I am still struggling with it.
13:39	tomjack	did you read the docstring?
13:42	tomjack	the in-transaction value you read by `@uid` may not be the same as the one committed by your `(commute uid inc)`
13:42	kaiyin	tomjack: I have read some posts on SO, now I seem to have some idea of how alter and commute differ.
13:44	kaiyin	alter will retry if some other thread came along and made changes during the transaction, while commute just commit the change.
15:00	fritzl	hey, is there a clojure function like haskells scan{l,r}?
15:03	TEttinger	(doc reductions)
15:03	clojurebot	"([f coll] [f init coll]); Returns a lazy seq of the intermediate values of the reduction (as per reduce) of coll by f, starting with init."
15:03	TEttinger	fritzl, I'm not sure if that's what you're after
15:03	TEttinger	,(reductions + (range 10))
15:03	clojurebot	(0 1 3 6 10 ...)
15:05	fritzl	TEttinger: that was what I was looking for, thanks
15:05	TEttinger	cool!
15:07	fritzl	kind of a better name as well
15:08	TEttinger	that's one of the things that I think clojure did a really good job on in the early design
15:13	mfikes	The names are good. I find the noun vs. verb aspect interesting. `reduce` or `reduction`
16:15	kaiyin	how do you add some file to classpath, after you add a new dependency in project.clj and run lein deps?
16:16	kaiyin	I mean before that the repl was already running.
16:17	noncom\|2	kaiyin: i think this is not supported by default (not recommended?)
16:17	noncom\|2	however, try pomegranate maybe
16:17	noncom\|2	or vinyasa
16:18	noncom\|2	i have some problem with lein: i run lein uberjar, but it keeps starting and then telling me, it cannot find some older version of java
16:18	noncom\|2	why lein tries to uberjar with a different java?
16:19	noncom\|2	how to fix?
16:26	kaiyin	noncom\|2: vinyasa looks cool. thanks.
17:07	kaiyin	how do you remove an var in repl?
17:12	kaiyin	for example (def lb (listbox :model (-> 'seesaw.core ns-publics keys sort))) and set up some listeners for the listview. after that this list just won't go away even though i redefine it to a number (def lb 3).
17:32	imanc_	hope clojure one day gets as popular as scala
18:02	ticking	imanc_: not going to happen, a huge part of scalas popularity comes from its java-esque-ness
18:03	imanc_	from what I gather, scala is a pretty mad language though and many don't like using it
18:07	ticking	imanc_: yeah, interestingly the new spark sql stuff from apache spark uses a pattern matching graph transformer and macro based code emitter, so it feels very clojurish in its implementation
18:31	sobel	imanc_: what makes it mad?
18:32	imanc_	sobel: overly complex in places, and many ways of doing things, plus the possibly to royally screw yourself wiht things like implicits
18:33	imanc_	apparently Odersky plans to simplify the language a fair bit
18:33	sobel	i am finding the composability of clojure addictive
18:33	sobel	and the metacoding tools are great.
18:34	imanc_	here's a scala core committer talking about why he binned scala and felt he wasted 5 years of his life working on it: https://www.youtube.com/watch?v=uiJycy6dFSQ
18:34	imanc_	i really like clojure - hope I can get a job working with it
18:36	sobel	imanc_: i took a java job that looked like a place i could shoehorn clojure into the mix
18:38	sobel	er..scAHla or scAYla?
18:38	imanc_	sobel - i'm in a similar situation. I have potential for a job that requires either scala or java 8, and I don't have commercial experience in either. But I'm thinking about trying to learn java quickly, and then seeing if I can introduce some clojure
18:38	imanc_	scaHla
18:38	sobel	i know enough java to wing it on a regular basis
18:39	sobel	especially including clojure usage
18:39	sobel	interop
18:40	sobel	well, now i have no desire to learn scala :)
18:40	imanc_	sobel: I've coded for years including some strongly typed langs, and the java lang looks pretty straight forward, but it's all the ecosystem and using it in an idiomatic way that takes time to get up to speed with. I'm wondering though, how long do you think i'td take to get to that sort of mid weight level with java?
18:41	imanc_	yeh, scala has some nice stuff, but it looks f'kin nasty in other ways.
18:41	sobel	if you know clojure but have to use java, learn to compose your implementations using Spring (XML or annotation, doesn't matter that much)
18:41	imanc_	'k
18:42	imanc_	annotatoin sounds preferable to xml
18:42	imanc_	:)
18:42	sobel	ymmv
18:42	sobel	i prefer xml because it is the most succinct way to yank all the creation code out of java
18:44	sobel	my generic-ish pattern was to replace important classes with interfaces (even eclipse can automate that refactoring) to make it easier to insert test impls/mocks with test versions of the xml object maps
18:44	imanc_	are you talking about IoC/dep injecion?
18:44	sobel	and to augment/replace most 'configuration' code with instructors
18:45	sobel	constructors
18:45	sobel	pretty much
18:45	imanc_	ah
18:45	sobel	closest thing to composition you get w/java :)
18:45	imanc_	yeh - going to suck going from python and clojure to java
19:00	xeqi	Bronsa: thanks for the formatting updates. No preview button, unknown formatting, and no easily seen edit button once created made me hope someone else would come along and fix it
19:02	Bronsa	xeqi: np, the preview button actually exists though, it's that blue icon next to the help one, underneath the textarea (took me a year to notice it)
19:04	xeqi	haha, so it does
19:08	brianpWins	Having some problems doing string replacemnts with escaped quotes: (string/replace (str "\"MARY\"") #"\"" #"") returns regex.Pattern cannot be cast to clojure.lang.IFn
19:53	sritchie	anyone know the elisp equivalent of “do"?
19:53	sritchie	can’t figure out how to write sequential forms
19:57	amalloy	progn
19:57	amalloy	sritchie:
19:57	sritchie	thanks
20:00	brianpWins	I still can’t find a way to do this string replacement: (string/replace "\"MARY\"" #"\"" #"") throws exceptions
20:01	amalloy	#"" is a regex, not a string. it doesn't make sense to replace with a regex
20:02	brianpWins	snap
20:02	brianpWins	i can’t believe I’ve been staring at this for like 2 hours lol
20:02	brianpWins	thanks amalloy
21:44	justin_smith	imanc_: what I have seen many places is people being able to convert ruby shops to use more and more clojure
22:46	myguidingstar	someone plz help explain why the two results here are different https://gist.github.com/myguidingstar/d7d2f03db35b2bcb7646
22:47	myguidingstar	(sorry about the function name, it's a simplified version)
23:57	mfikes	myguidingstar: Forcing realization of the sequence returned by index-symbols* man cause mutation of `result`. In the global case, the result of index-symbols* is not wrapped with dorun (so it is perhaps realized differently simply owing to printing at the REPL). If you wrap it with dorun, the results become the same.