#clojure log - Jun 26 2009

The Joy of Clojure
Main Clojure site
Google Group
IRC
List of all logged dates

0:12 codyK: any link to the technical explanation of the need to pre-declare identifiers?

0:19 http://groups.google.com/group/clojure/msg/c211f642141ca5a1

0:19 eh, looks like preference. wonder why that's not on clojure.org

1:35 Chouser: I still have to look up (doc nth) to see if the index or coll comes first

2:34 duncanm: Chouser: i'm lucky, i started using SLIME everywhere

2:35 Chouser: does that help?

2:36 duncanm: yeah, it shows the arglist in the minibuffer

2:36 Chouser: ah, nice.

3:25 Lau_of_DK: ,(filter #(= :a (key %)) {:a 5 :b 10})

3:25 clojurebot: ([:a 5])

3:27 Chouser: ,(select-keys {:a 5 :b 10} [:a])

3:27 clojurebot: {:a 5}

4:28 eevar2: http://www.icfpcontest.org/wordpress/?p=3

5:14 Ina: Woohoo, got my java callable clojure function working!

5:58 Jomyoot: Is clojure faster or slower than scala?

6:08 Chousuke: Jomyoot: That probably depends on what you're doing.

6:08 But I'm sure both are fast enough for most things.

6:08 Jomyoot: Making a web page

6:10 slashus2: oh yeah

6:27 Jomyoot: Does Clojure or Scala have larger community now?

6:30 combas: who can say, and it doesn't matter either way does it?

6:31 Jomyoot: how so?

6:31 larger community = more stuff

6:31 Chousuke: Keep in mind that Clojure is younger than scala.

6:32 arbscht: Jomyoot: you'll be disappointed to hear that neither community takes a census. :) also, I suspect they share many members.

6:32 clojurebot: They found no relationship between a programmer’s amount of experience and code quality or productivity.

6:32 Chousuke: what, is clojurebot sentient now? :P

6:32 combas: would you pick java over .net because java is on more mobile devices? or python over ruby because python is used by nasa? I hope not, pick the one you like and the one that works best for what you need to do, bigger doesn't always mean better

6:32 Jomyoot: is there a chance clojure will have mature web framework anytime soon?

6:32 Chousuke: I don't know about the Scala community, but Clojure people are generally very friendly.

6:33 arbscht: Jomyoot: by definition, no.

6:33 Jomyoot: why not?

6:33 Chousuke: Jomyoot: there's compojure and stuff.

6:33 arbscht: a mature web framework must take sufficient time to reach maturity!

6:35 combas: but if you want to count librarys then you should include java librarys when you count clojures, I've never touched scala so I really cant say anything about it

6:36 I've heard good things about scala though, not trying to knock it

6:47 mtd: I'm playing with clj-processing (cool) and I'm new to clojure, and I'm missing a newbie guide / tutorial to the REPL-time introspection features like python's help(obj) or CL's describe*. I've found doc, of course, but is there anything else?

6:48 perhaps just a help REPL-reflection idioms/tricks would be nice. Things like ways to use ns-publics, etc.

6:48 s/a help/pointers to/

6:56 achim: mtd: there's find-doc, and clojure.contrib.repl-utils has a couple of useful ones, like show (lists java methods of class/obj), source and expression-info

6:57 mtd: achim: thanks

6:57 achim: i don't know of any REPL interaction tutorial though

7:05 danlucraft: is there a REPL command to list functions in a module? (E.g. clojure.contrib.repl-utils)

7:13 mtd: danlucraft: I was expecting ns-publics to do what each "Public Variables and Functions" sections on http://code.google.com/p/clojure-contrib/wiki/OverviewOfContrib does, which is what I think you want, but it doesn't :(

7:15 achim: mtd: but it does ...

7:16 it returns a seq of [symbol var] pairs

7:16 ,(map first (ns-publics (find-ns 'clojure.core)))

7:16 clojurebot: (sorted-map read-line re-pattern keyword? val *compile-path* max-key list* ns-aliases the-ns == longs special-form-anchor instance? syntax-symbol-anchor format sequential? fn? empty dorun time remove-method gensym not= *3 unchecked-multiply doseq bit-or aset-byte if-not hash-set add-watch unchecked-dec some nil? string? second letfn keys for *2 long-array cond bit-set fn sorted? ns-unalias ns-publics dosync all-ns long wi

7:20 Jomyoot: Hi

7:21 danlucraft: achim, mtd: great! thanks

7:21 mtd: achim: ah sorry I was doing (ns-publics clojure.core)

7:21 achim: indeed, thanks

7:22 * mtd forgot the quote :(

7:22 achim: np. btw, *ns* represents the current namespace, no need for find-ns in that case

7:22 mtd: achim: yup, thanks.

7:22 Jomyoot: Why is clojure not in the programming language shootout?

7:38 cemerick: Jomyoot: no one's bothered to write code for it, I guess

8:50 rhickey: are refs eventually going to grow an :on-change slot (invoked on every set, not just at the close of a transaction like :validator)?

8:51 rhickey: cemerick: for what purpose?

8:51 Chouser: cemerick: that would be invoked synchronously inside the transaction I guess?

8:52 cemerick: Chouser: yes

8:54 rhickey: lots of different applications. Makes cells/dataflow stuff outrageously easy (as long as :on-change is triggered when a set applies a not= value, I'd think), makes redo-log-like stuff possible (for auditing, "replay", debugging, etc). The latter is possible with :validator, but oftentimes, visibility into in-flight state is very useful (as opposed to just the post-transaction value that :validator gets).

8:54 mmm, checkpointing

8:55 rhickey: not watchers?

8:58 doing dataflow inside transactions? must also be transactional or agent dispatch, else side-effect rules come into plaly

8:59 cemerick: rhickey: yeah, I've layered some stuff on top of refs to do what I need w.r.t. redo logging, it's all transactional.

8:59 I hadn't come across watchers before. I'm again floored by the depth of the waters.

8:59 rhickey: how scary should I consider "experimental" to be?

9:02 rhickey: at this point, not very scary, probably shouldn't be experimental anymore, I tagged them that way when I went back to synchronous watching, that seems to be most useful and accepted now. Say 'alpha', maybe a name change here or there

9:02 cemerick: so the watcher fns used to be tied to an agent or somesuch? (like add-watcher, vs. add-watch?)

9:03 rhickey: right, now all that stuff is uniform for all reference types

9:03 Chouser: are you going to keep both forms?

9:03 cemerick: rhickey: thanks, I'll dig into it.

9:04 rhickey: Chouser: dunno, the async can be built from sync, but, if kept, maybe should get a more explicit name

9:05 cemerick: yet another circumstance where I've thought "hey, I need XXX", and XXX is already in clojure or contrib. I really need to follow the mailing list more :-)

9:05 rhickey: http://github.com/richhickey/clojure/blob/0ac482878a1dd520cbee2faa0f5f6ab1082ffa76/src/jvm/clojure/lang/IRef.java

9:06 cemerick: yeah, I saw the *Watch methods, but for some reason, I thought it was part of the transaction machinery...

9:09 rhickey: at that point in the hierarchy, that puts those features on atom/ref/agent/var

9:09 * cemerick scrambles for Chouser's class hierarchy diagram :-)

10:03 rhickey: cemerick: check out tutorial.pdf, as mentioned here, it breaks down Clojure's Java interface abstractions and that graph

10:03 http://groups.google.com/group/clojure/msg/9b9559b8b092b7ee

10:13 cemerick: rhickey: heh, yeah, I was there :-) I've only recently been using refs significantly, so most of the time when you've talked about them in the past, though I've listened, it was sort of like listening to the teacher in Peanuts. ;-) Blub paradox applied to APIs, I suppose.

10:15 rhickey: cemerick: yes, that was less directed to you than to anyone looking at the monster graph, tutorial.pdf is much more accessible

10:15 also makes me seem like less of a crazy person

10:15 :)

10:15 Chouser: heh

10:15 cemerick: ha

10:15 Chouser: well that was my goal, so I'm upset you've thwarted me.

10:16 cemerick: well, I'm more of a grab-the-firehose, code-first-ask-questions-later sort of person, anywa.

10:16 anyway*

10:18 Put another way, I like falling into a hole a couple of times before learning how to build a bridge.

10:35 jackdempsey: hey guys, maybe dumb question but i've read some conflicting things and want to get this straight in my mind

10:35 clojure's speed was brought up on the mailing list recently

10:35 and i saw everything from its slow to its as fast as java

10:36 also that it compiles down to java code, and then ultimately bytecodes

10:36 so my naive assumption, just getting into clojure, is that ultimately once its bytecode, its all the same stuff, and speed shouldn't be an issue...

10:37 is that incorrect?

10:39 mblinn: so I think the standard advice is to make it work and then make it fast if you need to...

10:39 and in general, most things will probably not be as fast as Java unless you put some effort into it (type hinting, perhaps unsafe math ops, other fun stuff)

10:40 seems that the biggest win is usually getting rid of anyplace where clojure needs to use reflection

10:40 http://www.google.com/#hl=en&q=clojure+warn+on+reflect&aq=f&oq=&aqi=&fp=0k1C0PXk7Hc

10:41 whoops. meant http://clojure.org/java_interop

10:41 Chouser: jackdempsey: just because it's bytecode doesn't necessarily mean it's as efficient as the bytecode generated by some other compiler.

10:42 jackdempsey: or it could be faster. And how efficient the bytecode is depends not just on the compiler but on the code being compiled.

10:42 eevar2: jackdempsey: it's just clojure => byte code, there's no java code step

10:43 Chouser: and since the idiomtic way to solve one problem will often be different in Clojure vs. Java, it gets hard very quickly to use broad statements like "faster" or "slower".

10:44 eevar2: immutable data structures are likely to add some overhead tho. if you're not working on something paralellizable

10:44 jackdempsey: gotcha

10:45 ok so the clojure => java thing was false

10:45 wasn't sure how that would work...good to know :-

10:46 thanks all, appreciate it

10:46 coming from ruby, i deal with speed/performance questions all the time, and just wanted a bit of info to rely on here :-)

10:47 achim: clojure is fast enough to not care much about how fast it is ;)

10:47 jackdempsey: exactly

10:48 :-)

10:49 danlucraft: jackdempsey: I'm also coming from Ruby, and my perspective is that Java+Clojure are both so much faster than Ruby I don't care :)

10:50 jackdempsey: hehe, yeah

10:51 danlucraft: once you've learned to live with Ruby, you can live with anything

10:58 jackdempsey: heheh

13:51 jdm1: I have a question regarding metadata on functions, in particular running tests. I am following what the "Programming Clojure" book describes, and the docs as far as I can tell.

13:51 clojurebot: functions are maps

13:52 jdm1: I have a short snippet at http://pastebin.com/d59d41fa3

13:52 jackdempsey: hehe ty bot

13:52 jdm1: My question is why does it keep telling me "no-test" ?

13:54 JAS415: On june 26 at 13:52:47, clojurebot became self aware...

13:54 jackdempsey: hahah

13:54 Chouser: jdm1: (test #'sum)

13:54 jdm1: where in the book?

13:55 like, page number?

13:55 jdm1: page 267

13:55 they say (test a-var)

13:55 jackdempsey: um

13:55 isn't that the appendix?

13:55 Chousuke: jdm1: that's correct

13:55 (test sum) ; sum is not a var :)

13:56 #'sum is the var

13:56 jdm1: that worked, (test #'sum). It is in the section called "automating tests"

13:56 oh I see. I misread it completely

13:56 jackdempsey: yea 248 for me

13:57 jdm1: I am reading the pdf version of the book, that might be the difference in pages

13:57 jackdempsey: ah ok

13:57 jdm1: thanks, I don't know why I read that wrong, thanks for pointing that out.

13:58 weissj: i'm interested in running a clojurebot on our corporate irc server. anyone know how to set that up? i see hiredman you wrote it

14:01 jackdempsey: has anyone put together a list of various macro's like #' #^ and ^#'.....having a bit of a time trying to keep them straight in my head :-)

14:01 probably just need to use them a bunch

14:02 Chouser: http://clojure.org/reader

14:02 under Macro characters

14:03 jackdempsey: ah yes, ty

14:10 ah i see.....i was confusing meta with metadata

14:11 jwinter: Is there a way to do this with a functional literal: (map (fn [x] (rand-int 1000)) (range 1 10))

14:12 Chouser: ,(take 9 (repeatedly #(rand-int 1000)))

14:12 clojurebot: (39 289 522 137 423 135 471 633 588)

14:12 jwinter: that's much better, thanks Chouser

14:12 unlink: Is this really the correct syntax? http://dpaste.com/60140/

14:14 sh10151: Does anyone know of some good examples of using clojure.zip/xml-zip to edit XML?

14:14 I have a really simple transformation but I can't seem to wrap my head around using the library to do it

14:42 so I have xml like this: <a id="a" x="1" y="2"><b/></a>

14:43 how do I use the xml-zip to make that <a id="a" x="42" y="256"></b></a> ?

14:53 jackdempsey: man i really wish ctrl-d deleted in the repl

14:53 drewr: jline?

14:53 jackdempsey: bless you?

14:53 hehe

14:53 drewr: rlwrap?

14:54 jackdempsey: just used to emacs bindings on command line

14:54 unlink: use rlwrap

14:54 jackdempsey: ctrl-a, ctrl-e are nice

14:54 ok

14:54 off to google :-)

14:54 unlink: apt-get install rlwrap does it for me

14:54 jackdempsey: gotcha, os x here

14:54 unlink: then you can do rlwrap java clojure.main

14:55 though I recommend something like: http://dpaste.com/60165/

14:55 gripir: which ide/editors do you prefer with linux?

14:55 drewr: macports has both.

14:55 jackdempsey: cool

14:55 unlink: oh, you need to make a file called ~/.config/clj/completions...you can find those somewhere

14:57 ataggart: what's the difference between rlwrap and jline?

14:57 unlink: rlwrap has more features

14:57 and does more than just java

14:57 ataggart: ahh k

14:59 lisppaste8: Chouser pasted "edit with xml-zip, for sh10151" at http://paste.lisp.org/display/82572

15:01 jackdempsey: god, thats so nice

15:01 ty very much re: rlwrap

15:01 lisppaste8: Chouser annotated #82572 "...and all the way back to xml" at http://paste.lisp.org/display/82572#1

15:01 Chouser: sh10151: does that help?

15:04 Chousuke: I wonder if clojure.zip will be useful in manipulating the AST for c-in-c eventually :/

15:19 achim: is somebody working on a pure-clj reader?

15:19 Chouser: c-in-c will need that

15:20 unlink: Is there such a project?

15:20 Chousuke: not yet.

15:21 unlink: Then why is there a name besides "self-hosting compiler"?

15:21 Chouser: it will be more than the compiler

15:21 also reader, collections, etc.

15:21 Chousuke: well, I lied. there is a project. by chouser. but it's probably going to get scrapped when the real work starts.

15:22 I wonder if it'll really be worth it to write the clojure data structures themselves in clojure.

15:23 I suppose that would make them easier to port to non-JVM implementations.

15:25 Chouser: exactly

15:26 Chousuke: I broke my aquamacs somehow. It seems not to want to start up unless I set 'viper-mode to non-nil. :P

15:26 my emacs configs are a horrible mess anyway

15:26 I think I shall just remove them all and redo things from scratch

15:32 sh10151: Chouser: finally saw your paste -- thanks

15:32 I got something similar

15:32 Chouser: ok, good

15:32 sh10151: but emit seems to be doing funny things to whitespace

15:32 Chouser: yes

15:32 sh10151: unfortunately the other end is whitespace-sensitive

15:33 Chouser: contrib,lazy-xml has an emit that should leave it alone

15:33 sh10151: plug-in compatible?

15:33 Chouser: though the parse may discard some too :-/

15:33 yes

15:33 sh10151: discarding is OK I think

15:44 that did it -- thanks for the information

15:48 Chouser: sure

15:59 "git from the bottom up" is good.

16:00 one quick read and I'm already more effective

16:05 codefacekillah: if i have a struct s

16:05 is there an easy way to make (struct s ...) return a map with metadata attached to it?

16:05 like, somehow change the defstruct call?

16:06 Chouser: you don't just want your own fn that calls struct and then attaches some metadata before returning?

16:11 codefacekillah: i guess that would be easier

16:17 Chouser: if you really want it called 'struct' you can either use 'binding' to temprorarily set it thread-locally, or you can use your-namespace/struct instead of clojure.core/struct

16:21 slashus2_: http://blip.tv/file/2232410 A pretty interesting overview of the python GIL.

16:22 A disaster in its current state?

16:25 cemerick: slashus2: I presume jython doesn't have any such limitation (or pypy, for that matter)....

16:26 slashus2: I would assume that jython has the jvm's thread model.

16:26 I don't know about pypy

16:27 cemerick: given JRuby's perf of late, I'd presume that jython will follow. People who like those languages should have a decent MT environment for some time to come.

16:29 Chouser: tries to use deftype in c++

16:33 cemerick: Chouser: step away from the insanity :-)

16:35 Chouser: (deftype boost::unordered_PersistentMap<String .... wait a sec...

16:35 slashus2: Looks like the GIL falls on its face when dealing with multiple cores.

16:36 cemerick: yeah, it's been like that for a while, which is why GvR et al. are always pushing people towards fork, etc.

16:38 slashus2: Guido just put out a tweet with this link that I just posted. He hinted that he wants to fix the GIL for multi-core systems.

16:43 cemerick: heh, that's an epic change of direction

16:44 jackdempsey: yeah it is Chouser

16:47 unlink: GVR used a multi-core computer?

16:48 slashus2: unlink: What?

16:48 cemerick: unlink: ouch :-)

17:04 unlink: How do . and friends work with the .net clojure implementation?

17:22 Has any equivalent to if __name__ == '__main__' been developed yet?

17:26 Raynes: http://www.reddit.com/r/programming/comments/8w0l9/i_find_clojure_revolting/ This is the first time I've ever seen Clojure literally ran in the dirt before. :o

17:32 Chousuke: I think that rant passes as a definition of elitism.

17:32 slashus2: Seems like he doesn't realize that clojure will soon be written in itself.

17:32 dreish: I would call it high-powered cluelessness.

17:32 gnuvince_: Raynes: I stopped reading when the guy implied that Java libs are incorrect and that only something in pure Lisp is correct.

17:33 slashus2: I think I remember a guy commenting that argued along these lines. It may very well be this guy.

17:37 gnuvince_: slashus2: I wouldn't pay too much attention to the guy

17:38 slashus2: I don't pay very much attention to fringe groups unless their case is very compelling.

17:38 Especially people demanding paradigm shifts with no regard to practicality.

17:40 gbolyuba: hello

17:40 I have a question if I may

17:41 right after I do git clone git://github.com/richhickey/clojure.git I check for modifications with git status

17:41 and I see a several files with local modifications

17:42 $ git status

17:42 # On branch master

17:42 # Changed but not updated:

17:42 # (use "git add <file>..." to update what will be committed)

17:42 # (use "git checkout -- <file>..." to discard changes in working directory)

17:42 #

17:42 # modified: epl-v10.html

17:42 # modified: src/jvm/clojure/asm/AnnotationVisitor.java

17:42 # modified: src/jvm/clojure/asm/AnnotationWriter.java

17:42 # modified: src/jvm/clojure/asm/Attribute.java

17:42 # modified: src/jvm/clojure/asm/ByteVector.java

17:42 # modified: src/jvm/clojure/asm/ClassAdapter.java

17:42 # modified: src/jvm/clojure/asm/ClassReader.java

17:42 # modified: src/jvm/clojure/asm/ClassVisitor.java

17:42 # modified: src/jvm/clojure/asm/ClassWriter.java

17:42 # modified: src/jvm/clojure/asm/Edge.java

17:42 # modified: src/jvm/clojure/asm/FieldVisitor.java

17:42 # modified: src/jvm/clojure/asm/FieldWriter.java

17:42 # modified: src/jvm/clojure/asm/Frame.java

17:42 something like this

17:42 but I have just clonned the repo

17:42 can anyone tell me why this happens?

17:42 dnolen_: gbolyuba don't do that please

17:42 use a pasting service

17:42 gbolyuba: oh

17:42 ok

17:42 sorry

17:43 slashus2: oh my

17:44 gbolyuba: http://pastie.org/525954

17:44 full message

17:47 as far as I can see using git diff epl-v10.html, whole file shown as deleted and then same file is added

17:48 same with other files

17:48 Chousuke: :/

17:48 try git reset --hard origin/master

17:49 gbolyuba: HEAD is now at 5e34989 Test reading of numbers like "1." and "1.e3". Fixes #117

17:49 but message is the same

17:50 Chousuke: git status still shows them as modified?

17:50 weir.d

17:51 gbolyuba: yes

17:51 http://pastie.org/525963

17:51 here is output of git branches -a

17:51 achim: might there be some cr/lf conversion happening, depending on the platform?

17:51 gbolyuba: hmm

17:51 usuallygit gives a warning for that, no?

17:51 Chousuke: yeah, git shouldn't do that :/

17:51 gbolyuba: I did not get any warnings during clone

17:52 but it sounds reasonable if I think about it

17:52 let me check

17:53 Chousuke: is this on windows? I've never seen git do that

17:53 gbolyuba: it is

17:53 achim: i don't know much about git, but this is about the only reason i can think of why the diffs look like that

17:57 gbolyuba: well

17:57 I have FRLF

17:57 CRLF

17:57 even if I git checkout remotes/origin/master epl-v10.html

17:57 it still the same: modified in git status output

18:00 but eol sounds reasonable, gonna check on linux

18:01 btw, is I click IRC link on clojure.org (irc://irc.freenode.net/#clojure) I join ##clojure channel. Is it me doing something wrong?

18:01 slashus2: Try clojure in the url without the #

18:02 gbolyuba: that is what I did. I am here :)

18:05 ok, got my answer http://kerneltrap.org/mailarchive/git/2008/6/11/2093874

18:05 core.autocrlf false should work

18:05 achim: gbolyuba: i'm on OS X, in my checkout, the files that show up in the git status output you pasted have CRLF line breaks, the other ones have LF ...

18:05 ah, ok

18:05 gbolyuba: :)

18:06 let me try with core.autocrlf false

18:07 yep, git config --global core.autocrlf false and then clone repo works

18:07 thank you achim

18:41 eyeris: I have two files a.clj and b.clj. I defn a function c in b.clj. In a.clj I have (ns (:refer-clojure) (:use (b))), but Clojure complains the it can't resolve a call to c in a.clj. What am I doing wrong?

18:52 Chousuke: eyeris: you need to declare the namespace in b too, and it needs to be specified fully in 'ns as well.

18:52 eyeris: and ns requires the namespace name as its first argument anyway :)

18:53 eyeris: Yeah, I do both of those. I should have just pasted an example to pastebin.

18:57 I tracked it down.

18:58 I had the syntax wrong in a.clj and b.clj, but when I fixed it in a.clj, I had an identical error in b.clj and I just didn't notice the change in file name in the backtrace because it was scrolled off my terminal

19:39 sam_: Hello - I was wondering if anyone has been able to get a JavaFX program to be able to load up a clojure instance and then use it?

19:39 I'm currently trying that by creating a seperate java class that tries to load up a clj file and invoke it and then return the object to the JavaFX side

19:40 but RT.loadResourceScript can't seem to find my script in the classpath

19:40 and I'm not sure where it's looking exactly

19:50 Hello - I was wondering if anyone has been able to get a JavaFX program to be able to load up a clojure instance and then use it?

19:50 'm currently trying that by creating a seperate java class that tries to load up a clj file and invoke it and then return the object to the JavaFX side

19:50 but RT.loadResourceScript can't seem to find my script in the classpath

19:50 and I'm not sure where it's looking exactly

19:54 fffej: are any clojure people taking part in ICFP?

21:40 ataggart: what was the alternative to jline?

21:47 Chouser: rlwrap

23:20 is there not an alter-var-root for thread-local bindings?

23:34 combas: I prefer rlwrap, when I use jline (at least on linux) I dont get the pointer jump to the last bracket when I close a bracket, rlwrap also stores history between sessions

23:35 codefacekillah: in my slime session, i'm trying to do (add-classpath "file:///path/to/other/stuff.clj")

23:35 but my (ns ... (:use other.stuff)) is erroring out for some reason

23:35 combas: kind of makes me wonder though, why does the clojure repl need rlwrap or jline, why dont we just fix the repl so it works properly?

23:35 codefacekillah: and i've quadruple checked everything, so i'm not sure why it isn't working...

23:37 combas: I tried to add some example files to my class path the other day and it didnt work either, haven't figurd out whats wrong yet

23:40 codefacekillah: well, i guess i'll just add it to my $CLASSPATH then

23:40 that worked fine

23:40 but it'd be nice to not have to do that

23:42 weird...

23:42 it's still not working

23:42 ughghghg

23:43 combas: yeah when I added the files to my $CLASSPATH it didnt work either, I think the problem is that java doesnt check $CLASSPATH it checks something else, not sure what though

23:43 codefacekillah: oh whoops, i didn't add it to my ~/.emacs

23:46 combas: so you added something like export CLASSPATH=/path/stuff/ and then added (add-classpath "file:///path/stuff/file.clj") and it works?

23:58 ahhh.. i see what i did wrong now, i included part of the namespace of the files in the classpath.. doh

Logging service provided by n01se.net