00:06 (quit) Demosthenes: Read error: Connection reset by peer
00:06 (join) Demosthenes
00:40 (quit) jrslepak: Quit: Leaving
00:43 (quit) jeapostrophe: Ping timeout: 272 seconds
01:21 (quit) vkz: Quit: vkz
01:28 (quit) _AlbireoX: Ping timeout: 248 seconds
01:29 (nick) LeNsTR|off -> LeNsTR
01:39 (join) AlbireoX
01:45 (quit) yoklov: Quit: computer sleeping
02:14 (nick) chaozzbubi -> ChaozZBubi
02:20 (quit) Girffe: Quit: Page closed
02:37 (quit) jonrafkind: Ping timeout: 252 seconds
03:03 (quit) realitygrill_: Quit: realitygrill_
03:05 (join) realitygrill
03:09 (quit) carleastlund: Quit: carleastlund
03:23 (join) dzhus
03:28 (join) Blkt
03:30 Blkt: good morning everyone
03:36 (quit) gmcabrita: Read error: Operation timed out
03:40 (join) gmcabrita
03:41 (nick) gmcabrita -> Guest23552
03:43 (join) sindoc
03:47 (quit) EmmanuelOga: Ping timeout: 244 seconds
03:53 (quit) Shviller: Ping timeout: 276 seconds
03:54 (join) Shviller
03:55 (nick) ChaozZBubi -> chaozzbubi
03:55 (join) hkBst
03:58 hkBst: is anyone else having trouble searching the online manual?
04:41 (quit) realitygrill: Read error: Connection reset by peer
04:42 (join) realitygrill
04:43 (nick) LeNsTR -> LeNsTR|off
04:55 hogehoge: Hi. I have a windows executable (maybe) created with Racket's raco command. How could I extract the source code from the exe? I googled but the only thing I found was how to decompile bytecode. I'd like to know how to extract the bytecode from the exe.
04:59 (join) sid0
04:59 sid0: hiya, I had a question about macros and unbound identifiers
05:00 sid0: is there a way to treat an unbound identifier as a symbol, but only within a macro call?
05:01 sid0: I know I can override #%top in the module that defines the macro, but I'd like to restrict it to within macros defined by the module
05:02 sid0: so if foo is unbound (macro foo) shouldn't error out but (+ foo 1) should, etc
05:23 (join) noelw
05:58 (quit) realitygrill: Quit: realitygrill
06:03 (join) masm
06:09 (nick) Guest23552 -> gmcabrita
06:12 (part) sindoc
06:36 (quit) Demosthenes: Quit: leaving
07:22 (join) sindoc
07:32 hkBst: sid0: isn't that the default way it works of you put `foo' in the literals list? The problem with that is that the macro won't work in environments where this ``literal'' is not unbound. But you can use a keyword literal: #:foo
07:35 sid0: hkBst: I don't know the literals list beforehand
07:36 sid0: the unbound literal can be arbitrary -- though I now think it seems like a bad idea
07:38 hkBst: sid0: so then this literal (or symbol or keyword ?) would be bound by a variable?
07:42 sid0: hkBst: well, I wanted to bind any unbound literals to symbols
07:42 sid0: so if foo is unbound, it becomes 'foo
07:43 sid0: also, hm, does it make sense to attach contracts to macros?
07:47 aidy: Hi, is it possible to make fold-stream non blocking in this code, while only adding to (input-stream)?
07:50 (join) sindoc1
07:50 (quit) sindoc: Read error: Connection reset by peer
07:52 (join) jeapostrophe
08:00 (join) veer
08:13 (nick) chaozzbubi -> ChaozZBubi
08:14 (quit) veer: Quit: Leaving
08:32 (nick) ChaozZBubi -> chaozzbubi
08:54 (join) yoklov
08:55 (join) mceier
08:57 (quit) yoklov: Client Quit
08:59 (quit) jeapostrophe: Read error: Operation timed out
09:05 (quit) noelw: Quit: noelw
09:15 (nick) LeNsTR|off -> LeNsTR
09:18 (join) noelw
09:25 (quit) sindoc1: Quit: Leaving.
09:27 (join) EmmanuelOga
09:29 (nick) samth_away -> samth
09:32 (join) speps
09:33 (join) jonrafkind
09:37 (join) jeapostrophe
09:41 (join) dyoo
09:46 (join) jrslepak
09:59 (join) sindoc
10:06 (quit) jeapostrophe: Ping timeout: 244 seconds
10:18 (nick) chaozzbubi -> ChaozZBubi
10:31 (quit) jrslepak: Quit: This computer has gone to sleep
10:35 (quit) jonrafkind: Read error: Operation timed out
10:39 (quit) dzhus: Remote host closed the connection
10:39 (nick) jschuster_away -> jschuster
10:42 (quit) sindoc: Quit: Leaving.
10:45 (join) sindoc
11:05 aidy: forgot http://paste.lisp.org/display/127383 13:47 < aidy> Hi, is it possible to make fold-stream non blocking in this code, while only adding to (input-stream)?
11:11 (join) MayDaniel
11:14 hkBst: aidy: looks like you're trying to do something similar to functional reactive programming?
11:15 (quit) ASau: Ping timeout: 276 seconds
11:16 (quit) MayDaniel: 
11:22 aidy: hkBst: I guess so
11:23 hkBst: aidy: in that case, the stream you're trying to construct seems to be the wrong way round. New events should be added to the tail instead of the front. That way the logic of a stream agrees with the temporal order of the events you store in it.
11:24 (join) jeapostrophe
11:24 aidy: hkBst: they are?
11:25 hkBst: aidy: on second thought, maybe I misread your code
11:26 aidy: :)
11:27 aidy: I'll just write my own fold though
11:28 aidy: that way code on the other side doesn't even need to know about the event stream
11:29 (join) yoklov
11:31 hkBst: aidy: yes, I seems to me it would indeed be necessary to write your own fold which can examine the tail for an end-of stream-so-far marker or so...
11:43 (quit) sindoc: Read error: Connection timed out
11:43 (join) sindoc
11:49 (quit) hkBst: Quit: Konversation terminated!
11:50 (join) bluezenix
11:57 (quit) noelw: Quit: noelw
12:09 (join) bluezenix1
12:11 (quit) bluezenix1: Client Quit
12:12 (join) bluezenix1
12:12 (quit) bluezenix: Ping timeout: 260 seconds
12:16 aidy: http://paste.lisp.org/display/127394 does this look reasonable?
12:18 (join) sindoc1
12:18 (quit) sindoc: Read error: Connection reset by peer
12:22 (join) realitygrill
12:24 (quit) Blkt: Remote host closed the connection
12:25 (quit) sindoc1: Quit: Leaving.
12:32 (quit) yoklov: Quit: computer sleeping
12:34 (join) dzhus
12:39 RacketCommitBot: [racket] plt pushed 2 new commits to master: http://git.io/YT3YEQ
12:39 RacketCommitBot: [racket/master] fix bytecode compiler bug - Matthew Flatt
12:39 RacketCommitBot: [racket/master] add PLT_VALIDATE_COMPILE to enable extra checking of the compiler - Matthew Flatt
12:40 (join) ASau
12:45 (join) anRch
12:49 (join) yoklov
12:54 (quit) yoklov: Ping timeout: 255 seconds
13:04 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/_7J8eA
13:04 RacketCommitBot: [racket/master] racket/gui: fix refresh for right-aligned or centered paragraphs in text% - Matthew Flatt
13:12 (join) sindoc
13:14 (quit) anRch: Quit: anRch
13:15 (join) jonrafkind
13:16 (quit) sindoc: Client Quit
13:17 (join) realitygrill_
13:17 (quit) realitygrill: Ping timeout: 276 seconds
13:17 (nick) realitygrill_ -> realitygrill
13:19 (join) vkz
13:27 (nick) LeNsTR -> LeNsTR|off
13:41 (join) jrm
13:43 ASau: Stupid question: how do I reload file (with module)?
13:43 ASau: "require" seems to be load-once.
13:43 ianjneu: revert buffer
13:43 ASau: In command line :)
13:43 ianjneu: oh in the repl
13:43 ASau: Right.
13:44 ianjneu: I dunno. I always reload the repl, since I tend to have small interactions. Have you tried dynamic-require?
13:49 ASau: (Oh, my, and those people claim that Common Lisp is too complex...)
13:49 (quit) dzhus: Remote host closed the connection
13:52 vkz: guys, anyone here intimately familiar with HTDP textbook? I've worked through the 2nd edition all the way to Ch8 only to find it hadn't been finished. How frustrating :) Shall I just switch to HTDP 1st ed? Around part V seems reasonable? Any advice pls?
13:52 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/NDH6_Q
13:52 RacketCommitBot: [racket/master] The "libs" files are now under a "tags" subdirectory. - Eli Barzilay
13:52 ASau: Can I "delete-package" here?
13:53 ASau: Will it help?
13:54 jonrafkind: you mean like unrequire?
13:54 ianjneu: enter!
13:55 dyoo: vkz:  I think that's about right
13:55 ianjneu: Sayeth stamourv
13:57 (join) stamourv
13:57 ASau: jonrafkind: yes.
13:57 vkz: thx dyoo
13:57 dyoo: ASau: xrepl appears to have a mechanism for module reloading: http://docs.racket-lang.org/xrepl/index.html#(xrepl._require-reloadable)
13:58 dyoo: have to admit I haven't touched xrepl personally yet
13:59 ASau: Alright, if I "include" file, will it ignore module declaration?
13:59 ASau: Alternatively, if I remove module declarations, shall I be able to "include" it multiple times?
14:06 ASau: Alright, "include" works for me.
14:06 ASau: I wish I wouldn't be confused with modules.
14:09 speps: Hi guys, a racket build system question. With latest 5.2, in order to link against libracket3m.a on x86_64, I have to build racket with CFLAGS="-fPIC" or i get a known "could not read symbols: Bad value". Should this be pushed upstream?
14:09 ASau: You're doing something wrong.
14:10 dyoo: speps: that sounds like something; maybe send a ping to the racket-dev list
14:10 ASau: speps: if you're building a library, then you're doing it wrong.
14:11 ASau: I build on NetBSD with many sanity checks on and don't see this.
14:12 speps: ASau: I'm building fluxus from git
14:12 speps: http://www.pawfal.org/fluxus/
14:12 ASau: Does it build shared object?
14:13 speps: ASau: without building racket with -fPIC i get /usr/lib/libracket3m.a(salloc.o): relocation R_X86_64_TPOFF32 against `scheme_thread_locals' can not be used when making a shared object; recompile with -fPIC
14:13 ASau: I know that issue pretty well.
14:14 ASau: It appears when you build shared library and link against static one.
14:14 dyoo: Hmmm.. but it looks like a few others have been hitting this: https://bugs.archlinux.org/task/28119?project=1&openedfrom=-1+week
14:14 ASau: I see that Racket doesn't build DSO, it only builds static library.
14:15 ASau: I don't know what developers think about libtool.
14:15 ASau: Having Racket libtoolized would help.
14:15 speps: ASau: the case is exactly that a shared library linked against libracket3m.a
14:15 (quit) cdidd: Ping timeout: 276 seconds
14:16 ASau: speps: can you convert to static linking?
14:16 speps: dyoo: i opened that one
14:17 samth: ASau: racket already uses libtool
14:17 ASau: ?
14:17 ASau: Let me check it.
14:18 ASau: speps: wait a bit, racket build takes time.
14:20 speps: ASau: i have to check, btw seems like -fPIC is used by default yet for the racket garbage collector (gc) on x86_64, probably building the whole with fPIC is not that wrong
14:21 ASau: samth: I don't see how it uses libtool.
14:21 (join) patterngazer
14:21 ASau: samth: it seems to have some support but it is definitly incomplete.
14:21 ASau: At leats.
14:21 ASau: At least.
14:21 samth: ASau: incomplete in what sense?
14:22 speps: ASau: also fPIC build for static libs is a common practice http://www.gentoo.org/proj/en/base/amd64/howtos/index.xml?part=1&chap=3#doc_chap7
14:22 rudybot: http://tinyurl.com/743fvsk
14:22 ASau: salloc.@LTO@: $(srcdir)/salloc.c
14:22 ASau: 	$(CC) $(CFLAGS) -c $(srcdir)/salloc.c -o salloc.@LTO@
14:22 ASau: This should be $(LIBTOOL) --mode=compile --tag=CC $(CC) $(CFLAGS) ...
14:23 ASau: Or $(COMPILE.c) -o salloc.@LTO@ $(srcdir)/...
14:23 samth: ASau: i don't think we want to use libtool all the time
14:23 speps: probably i'll file a bug report
14:23 ASau: This code builds the library that causes problem as above.
14:25 (join) anRch
14:29 ASau: You can use suffix rule to avoid typing in this case.
14:33 (quit) patterngazer: Quit: OUCH!!!
14:33 (join) patterngazer
14:33 (quit) patterngazer: Client Quit
14:33 (join) patterngazer
14:33 (quit) patterngazer: Client Quit
14:34 (join) patterngazer
14:36 (quit) patterngazer: Client Quit
14:44 sid0: hm, syntax objects have lexical information stored along with them, right? in that case is there a way to basically get whatever an identifier in the lexical scope of a syntax object is bound to (at runtime)?
14:45 samth: sid0: lexical information is compile-time information, not runtime information
14:46 sid0: so inside a syntax-case template I guess
14:48 dyoo: sid0: so one thing you can do is non-hygienic capture, if I understand what you are asking.
14:48 chandler: sid0: I'm not sure I've completely understood your question, but it seems to me that you're mixing phases.
14:49 dyoo: so it's possible to do something evil like this:
14:49 sid0: dyoo: That's what I'm doing right now.
14:49 dyoo: #lang racket  (define-syntax (say-me stx)   (syntax-case stx ()     [(_)      (with-syntax ([me (datum->syntax stx 'me)])        (syntax/loc stx          (printf "I see me is: ~s\n" me)))]))  (let ([me "Danny"])   (say-me))
14:49 sid0: chandler: Yes, I think I am. Let me rethink
14:56 sid0: chandler: Yeah, on further reflection (har har) I think using a namespace will be a better way to go about this
14:58 (join) cdidd
15:13 (quit) bluezenix1: Quit: Leaving.
15:28 (quit) bfulgham: Read error: Connection reset by peer
15:28 (join) bfulgham
15:31 (quit) anRch: Quit: anRch
15:37 (part) jrm: "ERC Version 5.3 (IRC client for Emacs)"
15:43 (quit) vkz: Quit: vkz
15:48 (join) vkz
15:56 (join) dzhus
15:58 (join) yoklov
16:05 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/bwIATQ
16:05 RacketCommitBot: [racket/master] tighten bound for Mac OS X 10.6 localtime() hack - Matthew Flatt
16:09 sid0: chandler: thanks for that, btw. with namespaces the code becomes several times simpler
16:13 (quit) vkz: Quit: vkz
16:20 (join) dspt
16:34 samth: in which we are happy to have mflatt: http://venturebeat.com/2012/01/30/dahl-out-mike-drop/
16:34 (join) ticking
16:49 RacketCommitBot: [racket] plt pushed 1 new commit to release: http://git.io/hDMl5g
16:49 RacketCommitBot: [racket/release] tighten bound for Mac OS X 10.6 localtime() hack - Matthew Flatt
16:50 hogehoge: Does anyone know how bytecode (which can be decompiled with `raco decompile`) is stored in an exe created with raco?
17:00 samth: hogehoge: i believe it's basically just a literal byte string
17:01 (join) crest
17:02 crest: i just installed racket from freebsd ports and drracket crashs with a type error on start up "set method of mult-color%: expects type <real> as 1st argument, given: '1.0; other arguments were: '1.0 '1.0
17:02 crest: "
17:02 jonrafkind: what version
17:02 crest: 5.2
17:03 jonrafkind: well that sucks.. can you log it as a bug?
17:03 samth: crest: can you try an installer from http://racket-lang.org/download/
17:03 jonrafkind: bugs.racket-lang.org
17:03 jonrafkind: or try that first
17:03 samth: probably, this is a miscompilation or misconfiguration in freebsd ports
17:03 crest: samth: their is no install for freebsd (amd64)
17:03 crest: and i tried to build 5.2 from the unix sources
17:03 crest: same problem
17:04 samth: hmm
17:04 samth: if it's reproducible from source, then it is likely to be a bug in racket
17:04 samth: can you try a nightly build: http://pre.racket-lang.org/installers/ ?
17:05 crest: samth: i will build it as soon as possible
17:06 chandler: crest: which version of FreeBSD?
17:07 crest: chandler: FreeBSD 9.0/amd64
17:14 crest: waiting for make install
17:21 crest: still waiting for make install ...
17:22 crest: is it possible to run make install with multiple processes?
17:25 jonrafkind: its supposed to do so by default
17:35 hogehoge: samth: Sorry for my bad English. What I meant by "how bytecode is stored" is what kind of file format are used to store bytecode in a raco-created exe. What I want to do is to extract bytecode from a raco-created exe and to decompile it. To do so, I need to know how bytecode is stored. I thought bytecode was simply added to the end of the base(?) exe. But it seemed it was not. Because `raco decompile` returned "zo-parse: File too big."
17:37 crest: drracket version 5.2.1.3 start up
17:37 samth: crest: it should use multiple processors by default
17:38 samth: hogehoge: i don't know that there's an easy way to do that
17:38 samth: and even if it is at the end, raco decompile won't necessarily be able to handle it
17:38 crest: samth: it doesn't (maybe if i pass -j$N)
17:38 samth: because there's all the rest of the executable
17:39 samth: crest: the output of make install should say something like : raco setup: 9 making: <collection name here>
17:39 samth: with various different numbers in place of 9
17:44 (quit) ticking: Quit: Linkinus - http://linkinus.com
17:50 (join) ticking
17:53 (join) janpaulbultmann
17:53 (join) PfhorSlayer
17:53 (quit) cdidd: Quit: Leaving
17:54 (quit) ticking: Read error: Connection reset by peer
17:55 (quit) ivan`: Ping timeout: 252 seconds
17:55 PfhorSlayer: Is there a function that would make it easy to take an input string and return a list that is the input string split up by instances of a string I specify?
17:56 ianjneu: srfi/13 I believe has what you need
17:57 PfhorSlayer: I'll check that out, thanks!
17:57 ianjneu: string-tokenize
17:57 PfhorSlayer: that sounds right :)
17:58 PfhorSlayer: Actually, the input is already a path. I just want to split up the path and grab the name of the first directory
17:58 PfhorSlayer checks the path docs.
17:58 (quit) dyoo: Ping timeout: 245 seconds
17:58 PfhorSlayer: Yeah, explode-path should do what I want!
18:01 hogehoge: samth: I left out that I tried `raco decompile` after removing first some bytes (I used PEView to check the size of the base exe) to leave only what looks like bytecode. Anyway, thanks for your help.
18:01 ozzloy: i'm getting this error message when trying to build the stable branch: pastie.org/3285486
18:02 ozzloy: this is being build in a vm
18:02 ozzloy: make works, make install fails.  how do i whip it into shape?
18:02 samth: ozzloy: probably it's running out of memory
18:02 ozzloy: ah
18:02 (join) vkz
18:02 samth: what memory limit do you have on the vm?
18:03 ozzloy: 512MB
18:03 ozzloy: looking for where i can up it
18:03 (part) speps: "WeeChat 0.3.6"
18:03 (quit) mceier: Quit: leaving
18:03 ozzloy: found it.  i'll try upping to ... what do you think i should up it to?
18:03 ozzloy: 1G?
18:03 (join) ivan`
18:05 (quit) jrslepak_: Quit: Leaving
18:06 (join) jrslepak_
18:06 ozzloy: samth, ^
18:06 danlndi: ozzloy: i've upped mine, so up yours!
18:06 samth: i'd try that
18:06 ozzloy: ^_^
18:06 danlndi: i'll be going now
18:06 samth: i would also try running raco setup -j 1
18:06 samth: since you're using 4 processors, that will use more memory
18:07 ozzloy: initially the vm had 1 cpu.  i upped mine to see if that would help
18:07 samth: to do that, i'd run 'make plain-install'
18:07 samth: and then 'raco setup'
18:07 samth: or rather, 'raco setup -j 1
18:07 ozzloy: so ... raco is racket's make?
18:08 (quit) karswell__: Read error: Operation timed out
18:08 (join) karswell__
18:08 ozzloy: thanks, btw
18:09 (quit) dzhus: Remote host closed the connection
18:16 samth: ozzloy: raco is a general command line tool for racket
18:17 samth: 'raco make' is like 'make' in some ways
18:17 samth: 'raco setup' does a general setup of the standard library and other installed libraries
18:17 samth: other raco commands do other stuff
18:17 samth: (and raco is extensible)
18:20 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/fZrxZg
18:20 RacketCommitBot: [racket/master] adding gvector-remove-last! - Danny Yoo
18:25 janpaulbultmann: since when does drscheme have this (http://tinyurl.com/displayallthecode) sidebar, now I know where the bluej guys got the questionable idea from ^^
18:28 (quit) jeapostrophe: Read error: Operation timed out
18:28 samth: janpaulbultmann: ?
18:29 janpaulbultmann: the program contour ^^
18:30 janpaulbultmann: I mean DrRacket got it pretty much right, its hidable, the scaling factor is so small that it's really just a skyline, and it is not the main scrollbar. The guys from bluej (a java learning environment) adopted a similar thing, but it's unhidable almost readable and the main scroll bar ^^
18:31 janpaulbultmann: I always wondered where they got that weird idea from until I accidentally hit command U ^^
18:41 ozzloy: samth, upping to 1GB and reducing to 1 proc worked
18:41 samth: janpaulbultmann: ah, ok
18:41 samth: ozzloy: great
18:41 samth: ozzloy: the setup process can use a lot of memory
18:41 ozzloy: i just did make install
18:42 ozzloy: i'll have to look into raco more later
18:42 ozzloy: thanks again
18:42 samth: ozzloy: 'make install' is basically just doing the steps i described
18:42 ozzloy: cool cool
18:42 ozzloy: bbiaf
18:42 samth: it's just that doing them separately allows you control over number of processors used
18:49 (nick) jschuster -> jschuster_away
18:50 (quit) masm: Quit: Leaving.
19:09 (quit) yoklov: Quit: computer sleeping
19:10 (join) yoklov
19:14 (join) dnolen
19:15 (quit) realitygrill: Quit: realitygrill
19:30 (join) jeapostrophe
19:35 (quit) jeapostrophe: Ping timeout: 244 seconds
19:52 (quit) janpaulbultmann: Quit: Leaving...
19:53 (quit) jonrafkind: Ping timeout: 252 seconds
19:54 (join) ticking
20:06 (quit) yoklov: Quit: computer sleeping
20:15 (join) yoklov
20:16 (nick) samth -> samth_away
20:18 (join) Frozenlo`
20:21 (quit) Frozenlock: Ping timeout: 252 seconds
20:23 (quit) ticking: Quit: Linkinus - http://linkinus.com
20:25 (quit) Frozenlo`: Ping timeout: 260 seconds
20:29 (join) dan__
20:29 (nick) dan__ -> Guest21476
20:34 (part) HisDivineShadow: "I am gay for you"
20:43 (quit) Guest21476: Quit: Page closed
20:52 (join) jeapostrophe
20:54 (join) ticking
20:57 (quit) vkz: Quit: vkz
21:15 (join) jrslepak
21:23 (join) vkz
21:25 (quit) zerokarmaleft: *.net *.split
21:25 (quit) snorble: *.net *.split
21:25 (quit) eli: *.net *.split
21:25 (quit) jschuster_away: *.net *.split
21:25 (quit) ChaozZBubi: *.net *.split
21:25 (join) jschuster_away
21:25 (join) zerokarmaleft
21:26 (join) ChaozZBubi
21:28 (join) karswell
21:31 (quit) karswell__: Ping timeout: 276 seconds
21:31 (join) eli
21:40 (quit) dnolen: Ping timeout: 252 seconds
21:46 acarrico: 2.6 Locations: #%variable-reference talks about an anonymous variable reference, what the heck is that?
22:26 (nick) LeNsTR|off -> LeNsTR
22:27 (quit) vkz: Quit: vkz
22:33 (join) LeN00b2
22:34 LeN00b2: Hey all! Just wondering if someone could offer me a bit of insight. I'm trying to write a function that consumes an integer n and produces the shortest possible list of perfect squares that sum to that n. Lagrange proved that the any integer can be expressed as, at most, the sum of four squares. Any ideas?
22:36 LeN00b2: I've thought of a very brute-force approach but am wondering if there's something more elegent.
22:40 RacketCommitBot: [racket] plt pushed 1 new commit to release: http://git.io/cGrU5Q
22:40 RacketCommitBot: [racket/release] Move the dmg build to weatherwax (which can now deal with it. ) - Eli Barzilay
22:41 (join) dmj111
23:06 (quit) cataska: Ping timeout: 260 seconds
23:08 (join) cataska
23:11 LeN00b2: Anyone ? :D
23:14 (join) carleastlund
23:16 (join) jonrafkind
23:18 (join) realitygrill
23:18 LeN00b2: IRe-asking: Hey all! Just wondering if someone could offer me a bit of insight. I'm trying to write a function that consumes an integer n and produces the shortest possible list of perfect squares that sum to that n. Lagrange proved that the any integer can be expressed as, at most, the sum of four squares. Any ideas? I've thought of a very brute-force approach but am wondering if there's something more elegent.
23:21 jonrafkind: is that a project euler question?
23:21 (quit) jrslepak: Quit: Leaving
23:21 LeN00b2: No, it's a "my-dad" question. :) Showed him Project Euler a while back and he's been grilling me with all sorts of questions lately - this one seems very tricky.
23:31 (quit) LeN00b2: Remote host closed the connection
23:31 (join) veer
23:32 (join) LeN00b2
23:33 LeN00b2: Any ideas?
23:42 carleastlund: LeN00b2, no particular insight into it, myself.  Seems like a special case of the knapsack problem, which would mean an optimal solution has to be brute force (unless P=NP).
23:47 carleastlund: ...well, it suggests that, doesn't prove it.  It may be a special-enough case to be simpler.  Perhaps I just give up too easy :)
23:48 (quit) jeapostrophe: Ping timeout: 240 seconds
23:48 LeN00b2: Oh well. Brutre-force it is!
23:49 carleastlund: The shortest possible solution would probably be to do an "I'm feeling lucky!" Google search on "what is the shortest list of squares that adds up to <N>" and output the result.  ;)
23:49 jonrafkind: butar-force
23:50 LeN00b2: Haha. Yes, sadly, my dad would then own 50 bucks that used to be mine :P
23:50 LeN00b2: I've gotta stop making these bets!
23:50 jonrafkind: you should bet more manly things, like who can drink the most beer before passing out
23:51 LeN00b2: That's what my first undergrad was for, ten years ago!
23:51 jonrafkind: hell yea me too!
23:51 LeN00b2: Not that I remember all that much, but apparently I have a degree in Biology (And a very  grumpy ex-wife)
23:51 LeN00b2: :D