00:00 (quit) ambrosebs: Remote host closed the connection
00:00 (join) ambrosebs
00:01 (part) ambrosebs
00:07 (join) jeapostrophe
00:07 eli: ozzloy: That page confirms what I said about `regexp' having more history.  IMO, it's not softening that made `regex' go up, but the use of `regex' in some language (maybe perl).
00:08 eli: And the *sigh* was at Pauan, who insisted that it's not a string syntax.  (Something that became even clearer on the mailing list, BTW.)
00:11 (quit) jeapostrophe: Ping timeout: 244 seconds
00:13 (join) dnolen
00:16 (join) freakazoid
00:28 ozzloy: eli, oh it's totally expected that regexp would have more history.  but notice how its use starts going down?  it's not just "regex" going up
00:29 ozzloy: also larry wall is a linguist, so it's not surprising that he would use "regex" in perl
00:35 ozzloy: around 1900 there was a bump in use of regex and regular expression
00:36 ozzloy: just an odd thing i noticed
00:41 eli: ozzloy: Bah.  Being a linguist is irrelevant -- it's a programming language, an area where the even worse abbreviation "expr" is very popular.
00:42 eli: Also, see http://programmers.stackexchange.com/questions/102528/what-is-more-standard-shortening-for-regular-expression-regex-or-regexp/102569#102569
00:42 rudybot: http://tinyurl.com/7wf22vn
00:43 eli: So apparently English speaking people have no problems with an "xp" suffix.
00:46 ozzloy: eli, "r" is voiced, so "expr" would be preferable
00:47 ozzloy: and i believe "exp." being shorthand for "express" is for written communication, not spoken.  which i guess code mostly is anyways
00:48 mithos28: are we suggesting that regular expressions should be called regexprs?
00:49 ozzloy: i'm suggesting "regex" but not strongly.  i think at this point we're just debating for the sake of debating
00:49 ozzloy: we both already said it's not a big deal
00:49 ozzloy: but yes, regexpr would be slightly better than regexp for speaking
00:50 mithos28: I like my regexp, but thats probably from too much racket
00:53 ozzloy: yeah, i can see that
00:53 (join) jeapostrophe
01:06 (quit) jeapostrophe: Ping timeout: 260 seconds
01:07 freakazoid: I like pronouncing the abbreviation for expressway, "expwy"
01:16 ozzloy: freakazoid, ME TOO!  so my favorite!
01:16 freakazoid: haha
01:16 freakazoid: It's from reading the signs when I was a kid
01:17 freakazoid: I also read signs painted on the road backwards and say "Ped Xing"
01:17 freakazoid: My son's probably gonna grow up thinking that's how they're really pronounced.
01:17 ozzloy: well if that is how people pronounce them, then that is how they're really pronounced
01:18 ozzloy: i'm more on the descriptivist side of things
01:19 ozzloy: sort of a blend
01:23 (quit) francisl: Quit: francisl
01:40 (quit) attila_lendvai: Disconnected by services
01:41 (join) attila_lendvai
01:44 (quit) dnolen: Quit: dnolen
01:59 (quit) jrslepak: Quit: Leaving
02:05 (quit) freakazoid: Quit: Computer has gone to sleep.
02:29 (quit) jao: Ping timeout: 248 seconds
02:45 mithos28: What are standard ways to incrementally construct a recursive (self-referential) structure in an immutable style?
02:46 mithos28: I'm thinking about how to represent a grammar (BNF style) and allow changes to it
02:56 jonrafkind: use mutation and dont expose the mutators
02:57 (join) fridim_
02:57 mithos28: but that doesn't allow for an immutable incremental API
02:57 mithos28: without a lot of copying
02:58 jonrafkind: you mean you want users to be allowed to update the grammar?
02:58 mithos28: Yes
03:00 mithos28: I have had the same problem when trying to figure out how to represent graphs
03:02 mithos28: I have thought about doing it with state passing, which would work nicely in haskell with monadic do syntax. But in racket at that point it would be easier to do actual mutation
03:14 (quit) attila_lendvai: Ping timeout: 244 seconds
03:15 (quit) realitygrill: Quit: realitygrill
03:17 (quit) jonrafkind: Ping timeout: 240 seconds
03:28 (join) attila_lendvai
03:34 (join) bluezenix
03:37 (join) Blkt
03:40 (join) ahinki
03:40 (join) loz`
03:41 Blkt: good morning everyone
03:42 loz`: hi
03:44 loz`: i'm experimenting with gui in racket
03:44 loz`: but this doesn't work http://paste.lisp.org/display/126204
03:45 loz`: mouse events and paint callback are working
03:46 loz`: i'm not sure how to force canvas repainting
03:56 RacketCommitBot: [racket] plt pushed 4 new commits to master: http://git.io/2MmRIg
03:56 RacketCommitBot: [racket/master] Make the version tool use "Racket". - Eli Barzilay
03:56 RacketCommitBot: [racket/master] Make get-libs retry downloads if they look stuck. - Eli Barzilay
03:56 RacketCommitBot: [racket/master] Remove redundant execution bit from some files, convert one CRLF file to LFs. - Eli Barzilay
04:02 (quit) Shviller: Ping timeout: 252 seconds
04:02 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/PjeRMw
04:02 RacketCommitBot: [racket/master] New Racket version 5.2.0.5. - Eli Barzilay
04:02 (join) Shviller
04:09 (quit) noam: Ping timeout: 260 seconds
04:18 _p4bl0: loz`: you need to send the on-paint event, and also to set the color of the drawing context (seems to be white by default, which can be surprising)
04:18 _p4bl0: freaking paste.lisp.org *never* works for me :-/
04:18 _p4bl0: http://pastie.org/2948553
04:19 _p4bl0: loz`: other than that, it's a bad idea to redraw all the points each time, it will take longer and longer and longer, but I guess this is a small example just to show your problem
04:34 loz`: _p4bl0: ok, thanks)
04:45 loz`: it works :P
04:53 (quit) mithos28: Quit: mithos28
04:58 (quit) Pauan: Ping timeout: 252 seconds
05:14 (join) Pauan
05:15 (quit) bluezenix: Ping timeout: 244 seconds
05:19 (join) bluezenix
05:19 (quit) Pauan: Ping timeout: 252 seconds
05:23 (join) noam
05:26 (join) masm
05:33 (join) Pauan
06:07 (quit) noam: Read error: Connection reset by peer
06:07 (join) noam
06:13 (join) ahinki_
06:14 (quit) ahinki: Read error: Connection reset by peer
06:14 (nick) ahinki_ -> ahinki
06:22 (join) veer
06:22 (quit) Shviller: Ping timeout: 252 seconds
06:23 (join) Shviller
06:40 (quit) veer: Quit: Leaving
06:51 (join) ambrosebs
07:04 (join) attila_lendvai1
07:04 (quit) attila_lendvai: Disconnected by services
07:06 (quit) bluezenix: Quit: Leaving.
07:18 (join) haruki_zaemon
07:21 haruki_zaemon: Does racket have immutable data structures besides a cons list? eg, vector, hashmap/dictionary/associate array, etc?
07:31 Pauan: yes
07:32 haruki_zaemon: could u point me at them? Or are they all immutable by default?
07:32 Pauan: Racket has immutable conses, vectors, hash tables, and sets
07:33 Pauan: they're all in the reference:
07:33 Pauan: http://docs.racket-lang.org/reference/hashtables.html?q=hash%20table
07:33 Pauan: hash, hasheq, and hasheqv make immutable hash tables
07:33 Pauan: as does the procedures make-immutable-hash{eq, eqv}
07:33 haruki_zaemon: hmm ok.. I still see functions like hash-set!
07:34 haruki_zaemon: which is what I think threw me
07:34 Pauan: Racket has mutable *and* immutable hashes...
07:34 Pauan: hash-set! should only work on mutable hashes
07:34 Pauan: hash-set is functional
07:34 (join) bluezenix
07:35 Pauan: meaning it doesn't update the hash table: it returns a new one
07:35 haruki_zaemon: hmm ok so make-hash returns a mutable hash
07:35 haruki_zaemon: I need to use make-immutable-hash instead
07:36 haruki_zaemon: somewhat annoying but at least I know it can be done
07:36 Pauan: yes
07:36 haruki_zaemon: thanks
07:36 Pauan: hash-set! throws an error with immutable hashes, by the way
07:36 haruki_zaemon: right
07:36 Pauan: you can also use hash, hasheq, and hasheqv
07:36 Pauan: those return immutable hashes
07:36 haruki_zaemon: ahhh!
07:36 haruki_zaemon: right
07:36 haruki_zaemon: that's better
07:36 haruki_zaemon: nice
07:37 Pauan: (and since they're procedures, if you want to generate the key/value pairings dynamically, you can call apply on them with a list)
07:37 Pauan: (if you don't want to use make-immutable-hash)
07:37 Pauan: i.e. (apply hash '(foo "bar" qux "corge"))
07:38 haruki_zaemon: right
07:38 haruki_zaemon: ooh looks like they're also sequences
07:38 haruki_zaemon: nice!
07:38 Pauan: yup
07:40 haruki_zaemon: thanks
07:40 Pauan: sure
08:03 (join) veer
08:22 haruki_zaemon: hmm it's a shame -- thoiugh not a show stopper -- that the collections aren't named consistently :(
08:22 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/I-GgMw
08:22 RacketCommitBot: [racket/master] remove ? from German string constant to match similar removal from the English - Robby Findler
08:25 (quit) noelw: Quit: noelw
08:40 (join) jeapostrophe
08:48 (quit) veer: Quit: Leaving
08:50 (quit) ambrosebs: Remote host closed the connection
08:50 (join) ambrosebs
08:51 (part) ambrosebs
08:54 (nick) samth_away -> samth
08:56 samth: Pauan, haruki_zaemon: you can also use `for/hash' to construct hashes
09:23 (join) keenbug
10:04 danking: Did DrRacket drop the Add Teachpack menu item in the latest version?
10:04 danking: I've got 5.2.0.1--2011-10-10(dc912ee/a) and it's curiously missing from the Languages menu
10:06 (quit) haruki_zaemon: Ping timeout: 244 seconds
10:09 (join) realitygrill
10:11 samth: danking: you're probably in the wrong language level
10:17 (join) jrslepak
10:30 (quit) jrslepak: Quit: Leaving
10:46 (quit) ahinki: Quit: ChatZilla 0.9.87 [Firefox 9.0/20111122192043]
10:46 (join) dvanhorn
10:53 (join) freakazoid
10:59 (quit) realitygrill: Read error: Connection reset by peer
10:59 (join) realitygrill
11:03 (quit) jeapostrophe: Ping timeout: 244 seconds
11:06 (quit) realitygrill: Read error: Connection reset by peer
11:06 (join) realitygrill
11:09 (join) RackN00b
11:11 RackN00b: Hey uys - can someone comment as to how accurate drracket is when raising a two digit number to a large integer exponent? I'm writing a challenge to my coursemates to decrypt a certain message given a particular value but when I raise say 30 to the 337th power, I get a different answer from wolfram alpha... ideas?
11:12 bremner: is is floating point? all bets are off, basically
11:13 bremner: rudybot: init racket
11:13 rudybot: bremner: your sandbox is ready
11:13 RackN00b: Don't know much about floating point so I don't know what you mean :(
11:14 (quit) realitygrill: Ping timeout: 255 seconds
11:14 samth: RackN00b: if it's an integer (a whole number) then drracket is always exactly right
11:15 samth: that's also the case for fractions
11:15 RackN00b: nevermind - it seems to work - I was raising them to different power!
11:15 RackN00b: samth: Thanks... That's awesome.
11:15 Pauan: samth: I'm guessing that goes for all "exact" numbers, then?
11:15 samth: RackN00b: in more complicated settings, drracket sometimes produces inexact answers
11:15 samth: Pauan: yes
11:16 RackN00b: cool
11:16 samth: any arithmetic operation on exact numbers produces an exact answer
11:16 samth: rudybot: eval (exp 3 200)
11:16 rudybot: samth: your sandbox is ready
11:16 rudybot: samth: error: exp: expects 1 argument, given 2: 3 200
11:16 samth: rudybot: eval (expt 3 200)
11:16 rudybot: samth: ; Value: 265613988875874769338781322035779626829233452653394495974574961739092490901302182994384699044001
11:16 samth: rudybot: eval (/ (expt 3 200) 17)
11:16 rudybot: samth: ; Value: 265613988875874769338781322035779626829233452653394495974574961739092490901302182994384699044001/17
11:16 RackN00b: neat
11:26 (join) ambrosebs
11:27 (part) ambrosebs
11:27 RackN00b: How would I go about generating a list of random integers between 1 and 45 such that there are no repeats?
11:27 Pauan: hm
11:28 RackN00b: I need to basically take any character and create a numeric value for it.. Just wondering if there's a swifter method than writing it out...
11:28 Pauan: (shuffle (build-list 45 (lambda (x) (+ x 1))))
11:28 Pauan: that will give you a list from 1-45
11:28 Pauan: no number repeated
11:28 Pauan: in a random order
11:29 Pauan: oh!
11:29 bremner: then just take  prefix
11:29 Pauan: you can just use curry
11:29 RackN00b: argh! shuffle won't work in my language level :D
11:29 Pauan: drat
11:29 RackN00b: neither will curry unfortunately, but I can write my own version, in any case
11:29 Pauan: (shuffle (build-list 45 (curry + 1)))
11:29 Pauan: hm
11:30 Pauan: not sure how to do it without shuffle
11:30 Pauan: sans actually writing shuffle yourself
11:30 (quit) dvanhorn: Quit: dvanhorn
11:30 RackN00b: I don't think it would take me a reasonable amount of time... new to random, so...
11:30 Pauan: well
11:30 Pauan: you do have access to random, right?
11:30 RackN00b: How would I check? is there a random function?
11:30 Pauan: yes
11:31 Pauan: it's called "random"
11:31 RackN00b: what does it take are arguments?
11:31 Pauan: a number
11:31 Pauan: try (random 45)
11:31 RackN00b: works!
11:31 Pauan: okay
11:31 Pauan: I'll whip something up then
11:31 Pauan: but first
11:31 Pauan: do hash tables work?
11:31 Pauan: try (hash 'foo "bar")
11:32 RackN00b: Thanks Pauan... No idea - how do I check for hash tables (I don't even know what they are yet! ok)
11:32 RackN00b: not defined :(
11:32 Pauan: try this: (make-hash '((foo "bar")))
11:33 RackN00b: nope
11:33 Pauan: well, *dang*
11:33 samth: RackN00b: what language level are you using?
11:33 Pauan: that means this function will be *slow*
11:33 Pauan: (since it'll be recursing over a list, doncha know))
11:33 Pauan: though...
11:33 RackN00b: Intermediate Student with Lambda... that's ok - I only need to create it once
11:33 samth: also, here's a sketch of how to do it:
11:33 Pauan: ...do you have vectors?
11:33 RackN00b: nope
11:34 RackN00b: haha
11:34 samth: rudybot: (remove-duplicates (for/list ([i 20]) (+ 1 (random 45))))
11:34 rudybot: samth: ; Value: (45 13 39 4 5 18 26 3 27 25 9 10 31 40 43 19 33 41 7)
11:34 samth: rudybot: (remove-duplicates (for/list ([i 20]) (+ 1 (random 45))))
11:34 rudybot: samth: ; Value: (23 3 44 38 21 31 17 33 25 13 20 18 12 8 19)
11:34 samth: rudybot: (remove-duplicates (for/list ([i 20]) (+ 1 (random 45))))
11:34 rudybot: samth: ; Value: (40 35 44 39 18 36 5 11 33 10 4 29 9)
11:34 Pauan: but...
11:34 samth: RackN00b: it should be easy to write that loop with `build-list'
11:34 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/0I7mqg
11:34 RacketCommitBot: [racket/master] add a compiler-hint:cross-module-inline hint. - Robby Findler
11:34 Pauan: the list length varies then
11:34 Pauan: RackN00b, is the list supposed to be a fixed size?
11:34 samth: and you should be able to write `remove-duplicates' following the design recipe
11:35 RackN00b: Yes... I need to me able to map each english-language character to a number... Like an ascii map or something akin to that...
11:35 Pauan: okay then
11:35 Pauan: that won't work
11:35 Pauan: I'll whip something up
11:35 Pauan: using lists
11:35 Pauan: (vectors would be *much* faster, I think)
11:35 Pauan: (as would hash tables)
11:35 Pauan: (but at least it'll work)
11:35 samth: Pauan: for lists of this length, hash tables would probably be slower
11:35 RackN00b: We don't get to play with any cool stuff in cs 135 yet!
11:36 Pauan: samth: yeah, that's quite possible
11:36 Pauan: I'm just a big stickler for algorithmic complexity
11:36 Pauan: I see lists and I go "ewww linear"
11:36 Pauan: I see hashes and go "yay constant"
11:36 samth: there's a reason that fast sorting algorithms use bubble sort or insertion sort on short sequences
11:36 Pauan: yeah, I know sometimes a "worse" algorithm performs better
11:36 Pauan: due to various factors
11:37 samth: the first rule of performance is "Measure!"
11:37 Pauan: and when the size is small enough, almost any algorithm is just fine
11:37 ozzloy: hashes are not linear
11:37 samth: there is no second rule
11:37 ozzloy: er...
11:37 Pauan: ozzloy: correct
11:37 ozzloy: constant
11:37 Pauan: well
11:37 ozzloy: hashes are not constant
11:37 Pauan: it's not quite constant, no
11:37 ozzloy: they are linear
11:37 ozzloy: in the worst case
11:37 ozzloy: which is what big-o measures
11:37 Pauan: hm...
11:37 RackN00b: Can I get drracket to print a list of all characters it recognizes?
11:37 Pauan: docs don't say that
11:37 Pauan: they say O(log N)
11:38 Pauan: but N is limited, so it's effectively constant
11:38 Pauan: I wouldn't be surprised if hash tables can in very pathological cases end up being terrible performance, though
11:38 ozzloy: how big of a data set are we talking about here?
11:38 Pauan: in this case? about 45
11:39 ozzloy: the first rule of optimization is don't .... 45?!?
11:39 Pauan: haha
11:39 ozzloy: waste of time
11:39 freakazoid: Definitely need Hadoop for a dataset that big.
11:39 freakazoid: Speaking of which Facebook and Twitter are both looking for data people
11:39 Pauan: RackN00b: do you have assoc?
11:39 RackN00b: how do i check?
11:39 Pauan: well, I would just see what it evaluates too
11:39 Pauan: if it evaluates to a procedure, you're good
11:40 freakazoid: so if you know anyone for either send to seanl@literati.org (I actually work for FB but Twitter just pinged me on LinkedIn and I'm a nice guy)
11:40 RackN00b: also, I;m reading up the drracket documentation and there seems like there might be something that takes ... yep! bytes->list
11:40 (join) shofetim
11:40 RackN00b: Pauan:  not defined! haha
11:40 Pauan: well, dang
11:40 (quit) RackN00b: Remote host closed the connection
11:41 samth: ozzloy: yeah, you're right
11:41 samth: first rule: don't. second rule: measure.
11:41 (join) RackN00b
11:41 Pauan: what about member?
11:41 RackN00b: man.. colloquy keeps quittin on my... member? works
11:41 Pauan: okay
11:41 Pauan: that's all I need, then
11:41 (join) dnolen
11:41 samth: freakazoid: what do you work on at FB?
11:42 RackN00b: Pauan... wait - I might be able to use bytes->list
11:42 RackN00b: sorry.. list->bytes
11:42 Pauan: why did you ask for them to be random, then? :P
11:42 (quit) EmmanuelOga: Ping timeout: 248 seconds
11:42 RackN00b: Well they didn't absolutely need to be but I thought it would be nicer... but UTF-8 isn't in a particular order.
11:43 samth: RackN00b: utf8 is very much in a particular order
11:43 RackN00b: hmm..weird. so if A->65, B->66?
11:43 (join) dvanhorn
11:43 Pauan: there's a different code for lowercase and uppercase a, by the way
11:44 Pauan: but yes, B is 66
11:44 Pauan: in decimal
11:44 RackN00b: That's cool...Oh wait! no, it's not good. Others won't be able to use it - it has to be at intermediate student with lambda level if they want to tralate it back!
11:44 RackN00b: translate rather.
11:45 Pauan: aaaand bytes->list isn't defined?
11:45 RackN00b: nope :( I can't seem to find a list of defined functions organized per language level
11:45 Pauan: wait, why does it need to be random if you just want a pairing?
11:45 samth: RackN00b: see here: http://pre.racket-lang.org/docs/html/htdp-langs/intermediate-lam.html?q=intermediate
11:45 rudybot: http://tinyurl.com/82hbel7
11:45 Pauan: you could just do (build-list 45 (lambda (x) (+ x 1)))
11:46 RackN00b: Pauan - awesome, yep, that works...
11:46 Pauan: and then you could say that A is 1, B is 2, etc.
11:46 (quit) dvanhorn: Client Quit
11:46 RackN00b: Can that start at an integer other than 1
11:46 RackN00b: ?
11:46 Pauan: sure
11:46 Pauan: just add 1 to the lambda
11:46 RackN00b: cool
11:46 Pauan: (build-list 45 (lambda (x) (+ x 2))) causes it to go from 2-46
11:46 RackN00b: awesome, thanks
11:47 Pauan: build-list starts at 0
11:47 Pauan: hence the +1
11:47 (join) dvanhorn
11:47 RackN00b: samth: thanks for the link as well
11:47 (part) dvanhorn
11:47 (join) EmmanuelOga
11:52 RackN00b: (char->integer char) works! schweet
11:52 Pauan: oh, nice
11:52 (quit) dnolen: Quit: Page closed
11:52 (join) dnolen
11:53 (join) dingfeng
11:54 (join) anRch
11:56 dingfeng: (define xs (list 1 2 3)) (display (syntax-e #`(a #,@xs)))
11:57 dingfeng: (.#<syntax:c:\users\user\dropbox\racket-javascript\scratch.rkt:6:22 a> . .#<syntax (1 2 3 1 2 3)>)
11:57 dingfeng: why?
11:57 offby1: samth: geez
11:58 dingfeng: (define xs (list 1 2 3)) (display `(a ,@xs)) >>> (a 1 2 3)
12:01 (join) jeapostrophe
12:02 offby1: samth: thanks.  That, obviously, looks like some sorta buffering problem.  I swear I fixed something very much like this not long ago.
12:04 offby1: samth: https://github.com/offby1/rudybot/issues/11 if you care
12:05 (join) dvanhorn
12:05 dvanhorn: samth: hi
12:05 dvanhorn: rudybot: 'hi
12:05 (join) sigmat
12:05 rudybot: dvanhorn: your sandbox is ready
12:05 rudybot: dvanhorn: ; Value: hi
12:05 (join) mithos28
12:05 (join) jrslepak
12:05 (join) MayDaniel
12:06 dvanhorn: rudybot: (+ 3 2)
12:06 rudybot: dvanhorn: ; Value: 5
12:06 dvanhorn: rudybot: (define (! n) (cond [(zero? n) 1] [else (* n (! (sub1 n)))]))
12:07 rudybot: dvanhorn: Done.
12:07 dvanhorn: rudybot: (! 5)
12:07 rudybot: dvanhorn: ; Value: 120
12:07 dvanhorn: rudybot: (! 10)
12:07 rudybot: dvanhorn: ; Value: 3628800
12:07 dvanhorn: rudybot: (! 30)
12:07 rudybot: dvanhorn: ; Value: 265252859812191058636308480000000
12:08 jrslepak: rudybot: ((lambda (x) (list x (list (quote quote) x))) (quote (lambda (x) (list x (list (quote quote) x)))))
12:08 rudybot: jrslepak: ; Value: ((lambda (x) (list x (list (quote quote) x))) (quote (lambda (x) (list x (list (quote quote) x)))))
12:08 dvanhorn: rudybot: hey rudybot, can we go out for beers later?
12:08 rudybot: dvanhorn: It's not that we can't interpret it, it's that it's not the information we need.
12:09 (quit) bluezenix: Quit: Leaving.
12:09 (join) Olin_Shivers
12:09 Olin_Shivers: dvanhorn: HI
12:10 (part) RackN00b
12:10 sigmat: Olin_Shivers: suup
12:10 Olin_Shivers: dvanhorn: no I'm not
12:10 freakazoid: samth: At the moment I'm working on visibility into cache performance
12:10 (part) sigmat
12:10 (join) mechanik
12:10 Olin_Shivers: dvanhorn: I'm actually wearing my invisibility cloak. Look next to Martha
12:10 freakazoid: samth: high-density visualization of statistics, detecting and tracking down improper use of the cache, etc
12:11 jrslepak thought Chris was sitting further back
12:11 dvanhorn: rudybot: (define (f x) (f x))
12:11 rudybot: dvanhorn: Done.
12:11 dvanhorn: rudybot: (f 5)
12:11 rudybot: dvanhorn: error: with-limit: out of time
12:12 (join) sigmat
12:12 (part) dvanhorn
12:12 (part) sigmat
12:13 (quit) mechanik: Client Quit
12:13 (quit) Olin_Shivers: Client Quit
12:14 freakazoid: And beating front-end engineers over the head with sticks.
12:16 (quit) freakazoid: Quit: Computer has gone to sleep.
12:20 RacketCommitBot: [racket] plt pushed 5 new commits to master: http://git.io/vUh7WQ
12:20 RacketCommitBot: [racket/master] place scrbl fix - Kevin Tew
12:20 RacketCommitBot: [racket/master] Allow instanciation of places with resolved module paths - Kevin Tew
12:20 RacketCommitBot: [racket/master] Fix place.c typos - Kevin Tew
12:25 (quit) Blkt: Quit: ERC Version 5.3 (IRC client for Emacs)
12:29 ozzloy: samth, heh.  i've heard the second rule as "for experts only: don't do it *yet*"
12:33 ozzloy: Pauan, oh well then hash table doesn't do the naive thing when there's a collision, which i should have guessed
12:40 (quit) mithos28: Quit: mithos28
12:42 (quit) anRch: Quit: anRch
12:48 dingfeng: is unsyntax splicing supposed to behave the same way as unquote splicing?
12:49 samth: dingfeng: roughly, yes
12:49 samth: rudybot: #`(1 #,@(list 2 3))
12:49 rudybot: samth: ; Value: #<syntax:1:2 (1 2 3)>
12:50 dingfeng: rudybot: (syntax-e #`(1 #,@(list 2 3)))
12:50 rudybot: dingfeng: your sandbox is ready
12:50 rudybot: dingfeng: ; Value: (#<syntax:1:13 1> . #<syntax (2 3)>)
12:51 dingfeng: how does the dot come about?
12:51 dingfeng: rudybot: `(1 ,@(list 2 3))
12:51 rudybot: dingfeng: ; Value: (1 2 3)
12:52 (join) anRch
12:55 (join) jonrafkind
12:55 (join) ahinki
12:56 (join) kenjin
12:57 (nick) kenjin -> Guest30762
12:58 (quit) Guest30762: Client Quit
12:58 (join) kenjin2201
12:59 (quit) kenjin2201: Client Quit
13:00 (join) bluezenix
13:00 (join) RackN00b
13:00 RackN00b: Can anyone suggest how I should block together characters in my plaintext so that my algorithm doesn't produce what essentailly amount to a fancy substitution cipher?
13:04 jrslepak: if this just for playing around with simple crypto, try the Hill cipher
13:04 jrslepak: or did you already have a particular algorithm in mind?
13:05 RackN00b: I want to relate what we covered in math class to what we're doing in CS. I can encrypt character by character but that's kinda useless... It's RSA, I just can't see how to encrypt more than one character at a time...
13:05 jrslepak: oh ok
13:05 (quit) bluezenix: Quit: Leaving.
13:06 (quit) loz`: Remote host closed the connection
13:06 (join) loz`
13:06 jrslepak: for a simple example, consider your characters as digits (in base alphabet-size), decide how many digits to group together based on your RSA modulus
13:07 (join) freakazoid
13:08 RackN00b: Sorry I'm not clear on what you said: base alphabet-size ? And how does my RSA modulus affect the grouping?
13:08 jrslepak: 1. if there are n different characters available for writing messages in, think of each character in the message as a digit in base n
13:09 jrslepak: 2. you want there many possible plaintext blocks so that brute force is actually hard
13:11 RackN00b: ok... but I can arbitrarily choose how to block the palintext characters? If so, what does that block's numerical representation become? Ex, char1 -> 89 char2 -> 23... would I encrypt 8923 with my algorithm?
13:11 jrslepak: yes, if you've decided on 2-character blocks
13:12 RackN00b: So what if your character blocks don't divide the number of characters in your plaintext?
13:13 jrslepak: you could pick a "null" character to fill up the rest of your final block
13:15 RackN00b: Coo.. Now's I've converted my plaintext to the ascii codes - what if some of them are three digits and some are two digits?
13:16 jrslepak: 65 --> 065
13:16 (quit) anRch: Read error: Connection reset by peer
13:16 jrslepak: (if that's how you want to encode them...)
13:16 RackN00b: I see... Is there any way in DrRacket to combine two numbers in such a manner? ie.  (somefoo 23 24) --> 2324
13:17 jrslepak: might be easier to say (65,106,39) -> 65*128^2 + 106*128^1 + 39*128^0
13:17 dingfeng: could anyone explain to me how the sexpr "(syntax-e #`(1 #,@(list 2 3)))" produces a dotted list?
13:18 RackN00b: I see. thanks - Imma see if I can get this working thanks
13:18 jrslepak: but the combination you ask for... look into number->string and string-append
13:18 RackN00b: Will do, thanks
13:21 (quit) keenbug: Ping timeout: 244 seconds
13:25 (quit) acarrico: Read error: Operation timed out
13:26 asumu: Anyone know offhand of a way to use 2htdp/image images as picts?
13:28 (quit) jrslepak: Quit: This computer has gone to sleep
13:31 stamourv: dingfeng: It doesn't. It's just that `syntax-e' only "peels off" the outermost syntax layer, exposing a pair.
13:31 stamourv: If you were to peel off more, you would get a list.
13:31 stamourv: rudybot: (syntax->datum #`(1 #,@(list 2 3)))
13:31 rudybot: stamourv: your sandbox is ready
13:31 rudybot: stamourv: ; Value: (1 2 3)
13:31 stamourv: rudybot: (syntax->list #`(1 #,@(list 2 3)))
13:31 rudybot: stamourv: ; Value: (#<syntax:1:17 1> #<syntax 2> #<syntax 3>)
13:39 (join) acarrico
13:40 dingfeng: but are they the same, given that i can't remove the syntax when doing macros?
13:40 dingfeng: my macros fail unless i
13:40 dingfeng: do something like #,(append (list #'j-var) args)
13:42 stamourv: You're right that you don't want to peel off too much, but `syntax->list' is usually fine.
13:42 stamourv: And since you actually care about the list structure in your macro, it makes perfect sense to expose it.
13:42 (join) dingfeng_
13:45 (quit) dingfeng: Ping timeout: 265 seconds
13:45 (join) dvanhorn
13:48 RackN00b: Argh! How do I keep leading zero on an integer? I'm trying to express 25 as 025 and I do a string-append to the converted number 25, but when I convert back, she switches back to 25... clues?
13:48 Pauan: that's because the reader strips it away
13:49 Pauan: you'll have to keep it in string form if you want the leading 0
13:49 Pauan: for the same reason that "25.50" turns into 25.5
13:51 ozzloy: is there a better way to get the first value out of (values 1 2) than to do (let-values ([(a b)]) (values 1 2) a)?
13:51 (quit) MayDaniel: Read error: Connection reset by peer
13:51 ozzloy: rudybot, eval (first (values 1 2))
13:51 rudybot: ozzloy: your sandbox is ready
13:51 rudybot: ozzloy: error: context expected 1 value, received 2 values: 1 2
13:52 ozzloy: rudybot, eval (let-values ([(a b) (values 1 2)]) a)
13:52 rudybot: ozzloy: ; Value: 1
13:52 ozzloy: (pretend i got the brackets right the first time)
13:52 Pauan: there's "call-with-values", but...
13:52 RackN00b: Crap.. I can't do any arithmetic on it if it's in string form...
13:52 Pauan: not sure if that's really any "better"
13:53 Pauan: RackN00b: that's correct, you'll need to convert it into a string to do math
13:53 Pauan: in which case you'll probably want to save the "convert to string and pad it" bit for last
13:53 RackN00b: Oh wait.. I can just append something other than 0... it should still work...
13:53 Pauan: or at least as late as you can
13:58 dingfeng_: well, i still don't see why the dotted list is generated... is that the inteded behavior? doesn't quite make sense. and it does make a difference when handling symbols from other contexts
13:58 ozzloy: i was going to suggest format, but it looks like it doesn't have the equivalent of c-style %03d
14:04 RackN00b: Is there  a way to test for an empty string as a base case for recursion through a string? I want to spilt a string into 16 character blocks...
14:04 Pauan: I guess you could use (equal? x "")
14:04 RackN00b: cool
14:09 (quit) metadave: Read error: Connection reset by peer
14:09 (join) metadave
14:21 (quit) attila_lendvai1: Ping timeout: 248 seconds
14:22 (quit) dingfeng_: Ping timeout: 265 seconds
14:25 (join) MayDaniel
14:30 (join) mithos28
14:31 stamourv: In Racket from up to date git, I can't get the macro stepper to start. Anybody has the same issue?
14:32 stamourv: Actually, that seems to be only for TR files.
14:32 (join) anRch
14:33 stamourv: Upon further inspection, it's only for TR files that don't typecheck.
14:37 (quit) MayDaniel: Read error: Connection reset by peer
14:38 jonrafkind: stamourv, are you running from drracket?
14:38 jonrafkind: try this on the command line: racket -e "(require macro-debugger/stepper) (expand-module/step \"$1\")", and replace $1 with a filename
14:38 jonrafkind: i just put that into a file called 'macro-stepper'
14:41 (join) jschuster
14:45 RacketCommitBot: [racket] plt pushed 2 new commits to master: http://git.io/nWyS7w
14:45 RacketCommitBot: [racket/master] add a contract for slideshow/pict's bitmap function - Robby Findler
14:45 RacketCommitBot: [racket/master] drop the weekend condition for the default icon style - Robby Findler
15:17 (join) Blkt
15:41 asumu: Hooray to whomever wrote the shadow-frame function in unstable for slideshow.
15:41 asumu: pslides + shadow-frame + a nice font (like Cantarell) => nice looking slideshow presentations
15:41 jonrafkind: can you show a quick example?
15:42 jonrafkind: whats pslides
15:43 asumu: Allows for "drag and drop" style slides rather than the append together style of normal slideshow slides.
15:44 asumu: Used like (pslide #:go (coord 0.3 0.3) (t "Upper left") #:go (coord 0.7 0.7) "Down here")
15:44 jonrafkind: where do you get it, i dont see it in the docs
15:44 asumu: jonrafkind: http://pre.racket-lang.org/docs/html/unstable/ppict.html
15:45 jonrafkind: oh tags look nice
15:46 (quit) anRch: Quit: anRch
15:47 ozzloy: anyone in here taking the ai-class.com course?
15:48 (quit) ahinki: Quit: ChatZilla 0.9.87 [Firefox 7.0.1/20110928224103]
15:49 asumu: ozzloy: Wanted to, but had no time. :\
15:49 asumu: (incidentally, someone should do a pl-class.com)
15:50 ozzloy: asumu, it's a cool class.  actually i want the answer to be "no" because i want to ask about my implementation and not worry about academic integrity rules
16:18 RacketCommitBot: [racket] plt pushed 2 new commits to master: http://git.io/k81jEg
16:18 RacketCommitBot: [racket/master] fix variables assigned but not referenced - Kevin Tew
16:18 RacketCommitBot: [racket/master] fix bug in cross-module inliner - Matthew Flatt
16:19 ozzloy: what's a good pastebin for racket code?
16:19 ozzloy: oh maybe gist
16:21 ozzloy: i have implemented the maxValue algorithm shown at 1:31 here http://goo.gl/O5gZ8 using this code: https://gist.github.com/1419964 i'm hoping someone can stop laughing at it for long enough to tell me how to do it better
16:21 samth: ozzloy: also paste.lisp.org
16:23 ozzloy: where better is determined by some function that values idiomatic racket and fewer sexps, possibly without the initialization to inf.0
16:26 samth: ozzloy: https://gist.github.com/1419997 is a little nicer, maybe
16:26 samth: you might just write it with a loop
16:26 samth: instead of a `for'
16:30 (quit) bfulgham: Ping timeout: 245 seconds
16:30 ozzloy: samth, thanks
16:31 stamourv: jonrafkind: Yeah, that's from DrRacket, lemme try your thing.
16:32 stamourv: Oh, that uses the text stepper, right?
16:32 (quit) metadave: Read error: Connection reset by peer
16:32 stamourv: xrepl can do that already.
16:32 (join) metadave
16:32 stamourv: But I want to explore the syntax objects, which the text version can't do, IIRC.
16:33 (part) RackN00b
16:34 ozzloy: samth, would your version stop after the first successor-value that surpasses the upper-bound, or skip all the iterations where successor-value is greater than upper-bound?
16:34 jonrafkind: stamourv, it spawns the gui
16:36 stamourv: Oh, does it? I'll give it a try then.
16:37 samth: ozzloy: skip them
16:39 ozzloy: samth, the original algorithm shows returning maximum-value after the first that exceeds the upper-bound, beta
16:39 samth: ah, ok
16:39 samth: yours iterates over everything, but just doesn't do anything for those values?
16:39 ozzloy: yeah
16:39 ozzloy: doesn't descend
16:40 ozzloy: i was thinking for/first would be what i wanted
16:40 ozzloy: but i couldn't wrap my head around how to update the v and alpha
16:47 samth: you could just change my code or yours to use `for/first'
16:49 (join) bfulgham
16:50 stamourv: jonrafkind: That works, thanks!
16:51 jonrafkind: nice
16:51 stamourv: But it only works with relative paths.
16:51 jonrafkind: meaning what
16:51 (join) keenbug
16:51 stamourv: You can't have an absolute path as your $1.
16:52 jonrafkind: oh interesting.. i wonder why that is
16:52 stamourv: `expand-module/step' takes a module-path.
16:52 stamourv: And absolute paths are not in the grammar for module-path.
16:53 samth: stamourv: you can use (file "/absolute/path")
16:53 stamourv: Ah, good point.
16:57 stamourv: eli: What do you think about adding jonrafkind's trick to launch the graphical macro stepper to xrepl?
16:59 jonrafkind: does xrepl have a way to add user commands yet? i forget
17:02 (quit) dvanhorn: Quit: dvanhorn
17:04 (join) dvanhorn
17:09 (join) bluezenix
17:10 (join) leandrosansilva
17:11 (join) RackN00b
17:13 RackN00b: How would I create a list of possible permutations of three symbols... I ask for the following reason: I have three functions that act on a particular structure. Each of those three functions can be called three times in a given "run". So the max number of function calls is 9. I would like to create a list that has all combinations of those functions up to 9... any ideas?
17:15 samth: RackN00b: there's an htdp problem about permutations
17:15 samth: problem 12.4.2
17:15 samth: you should look at it
17:15 RackN00b: Ooh.. I shall take a look - thanks
17:15 samth: and follow the design recipe to solve it
17:15 samth: really, trust me on that one
17:15 RackN00b: will do, thanks
17:15 RackN00b: Can I apply my powerset function for this?
17:16 stamourv: samth: Found the problem with `for/sum'.
17:16 (join) kbs
17:16 samth: stamourv: awesome
17:16 stamourv: Look at the expansion of `(for/sum: : Float ((i : Float (quote (1.2 3.2)))) i)'.
17:16 samth: is it that it's generalizing 0 to Number?
17:16 stamourv: No.
17:16 samth: i mean integer?
17:16 stamourv: Yes.
17:16 samth: right
17:16 samth: but it doesn't work even if you change it to `Real'
17:17 stamourv: That means that the argument is now Integer.
17:17 stamourv: And Floats don't work there.
17:17 samth: so we need to be groveling more to find the relevant information
17:17 stamourv: Yes. I said I found the problem, not a solution.
17:18 samth: fixing `for/sum:' with `Real' as the annotation should be pretty easy
17:18 samth: but `Float' is harder
17:18 stamourv: I don't see how.
17:18 stamourv: We can't inject the annotation.
17:18 samth: why not?
17:18 stamourv: It's too deep in the expansion.
17:18 samth: just do the expansion yourself
17:18 stamourv: Same problem we had with `for/hash' and other.
17:19 stamourv: Wait, you mean have the implementation of `for/sum:' expand `for/sum', then look through the remains?
17:19 samth: (for/sum seqs . body) = (for/fold ([i 0]) seqs . (let ([v (let () . body)]) (+ v i)))
17:19 stamourv: Didn't we decide that that was a horrible solution?
17:19 stamourv: Oh, that/.
17:19 stamourv: Yeah, can do.\
17:19 samth: just implement it yourself
17:20 stamourv: In terms of `for/fold' makes sense.
17:20 samth: since it's simple and doesn't depend on any of the hairy parts of `for'
17:20 shofetim: I am working on a presentation in #lang slideshow, editing in emacs and sending 'slides' as I build them to a repl running in emacs. But when I make a syntax error the slideshow closes, and I have to restart the repl to get it back...
17:20 samth: which we don't want to touch
17:20 stamourv: Right, I thought you meant reimplementing the whole thing.
17:20 samth: shofetim: yeah, there's some mutable state in `slideshow' which can only be used once
17:20 samth: it can be annoying
17:21 shofetim: so I have to reload from the beginning and navigate back to the slide I was on?
17:22 (quit) kbs: Quit: leaving
17:27 samth: shofetim: check out `start-at-recent-slide'
17:28 shofetim: samth: nice : )
17:37 stamourv: samth: Got the Real part working.
17:37 samth: stamourv: awesome
17:37 stamourv: Now, the question is: Calling `subtype' from inside `base-env/prims.rkt': ugly or disgusting?
17:37 stamourv: If it's merely ugly, I can fix the Float part.
17:39 samth: disgusting
17:40 stamourv: Ok, then I got nuthin'.
17:41 samth: yeah, me neither
17:41 samth: the problem is this:
17:42 samth: (for/sum: : Float ([i : Float '(1.1)] #:when (zero? (random 1))) i)
17:43 stamourv: Well, my disgusting solution would have fixed that.
17:43 samth: by changing the semantics of for/sum: when you annotate it with Float?
17:43 stamourv: Yes.
17:43 samth: that's worse than disgusting
17:44 stamourv: But good point, that's a change of semantics, so that's out.
17:44 samth: maybe there should be an flsum
17:44 samth: rather, for/flsum
17:44 stamourv: Bleh.
17:44 stamourv: Would only be useful for us.
17:45 samth: well, you can get dynamic errors in racket b/c of this
17:45 samth: (flvector-set! fv 0 (for/sum ...))
17:47 stamourv: Yes, but only if you're summing over empty sequences.
17:47 stamourv: But good point.
17:48 stamourv: Numbers are hard, let
17:48 stamourv: 's do XML!
17:48 samth: how much do you want to know about character sets?
17:49 (quit) Blkt: Remote host closed the connection
17:50 stamourv: All the world is ASCII?
17:50 stamourv flees.
17:50 samth: what was your name again?
17:51 stamourv: All ASCII.
17:51 stamourv: As problematics as dashes are, they're ASCII.
18:15 (quit) jeapostrophe: Ping timeout: 255 seconds
18:20 shofetim: How do I make a pict fill an entire slide? (no border etc)?
18:21 samth: what do you mean?
18:21 samth: i haven't had a problem doing that
18:22 shofetim: I don't see anything in the docs (I may be missing it). I have an image that is roughly 1200x1000 and I want it to be the slide... no border etc. Right now I can get it to fill up everything but about a 20px wide border
18:22 samth: shofetim: try using the `inset' function with -20
18:23 samth: see also set-margin!
18:25 shofetim: samth: thanks : )
18:31 (quit) leandrosansilva: Ping timeout: 260 seconds
18:46 (join) jeapostrophe
18:54 (quit) bluezenix: Quit: Leaving.
19:02 (quit) dnolen: Quit: Page closed
19:03 (quit) dvanhorn: Quit: dvanhorn
19:04 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/22G3OA
19:04 RacketCommitBot: [racket/master] Use `bitmap' instead of `icon->pict' - Neil Toronto
19:12 (quit) keenbug: Ping timeout: 252 seconds
19:55 (quit) masm: Quit: Leaving.
19:59 (join) dvanhorn
20:02 (nick) samth -> samth_away
20:03 (join) francisl
20:04 (quit) freakazoid: Quit: Computer has gone to sleep.
20:08 (join) freakazoid
20:17 (join) kingless
20:21 (part) RackN00b
20:23 (quit) francisl: Remote host closed the connection
20:24 (join) francisl
20:31 (quit) fridim_: Ping timeout: 248 seconds
20:32 (quit) shofetim: Ping timeout: 255 seconds
20:32 (quit) jonrafkind: Ping timeout: 240 seconds
20:40 (quit) kingless: Quit: Colloquy for iPhone - http://colloquy.mobi
20:40 (join) dingfeng
20:45 dingfeng: rudybot: (define-syntax foo (lambda (stx) (print (syntax-e stx)) #'(display "\n"))) (foo 1 2 3)
20:45 rudybot: dingfeng: could anyone explain to me how the sexpr "(syntax-e #`(1 #,@(list 2 3)))" produces a dotted list?
20:46 dingfeng: rudybot: (print (syntax-e #`(foo 1 #,@(list 2 3))))
20:46 rudybot: dingfeng: your sandbox is ready
20:46 rudybot: dingfeng: ; stdout: "'(#<syntax:1:20 foo> #<syntax:1:24 1> . #<syntax (2 3)>)"
20:46 dingfeng: rudybot: (begin (define-syntax foo (lambda (stx) (print (syntax-e stx)) #'(display "\n"))) (foo 1 2 3))
20:46 rudybot: dingfeng: ; stdout: "'(#<syntax:1:83 foo> #<syntax:1:87 1> #<syntax:1:89 2> #<syntax:1:91 3>)\n"
21:13 freakazoid: rudybot: (syntax-e #`(foo 1 #,@(list 2 3) #,@(list 4 5)))
21:13 rudybot: freakazoid: your sandbox is ready
21:13 rudybot: freakazoid: ; Value: (#<syntax:1:13 foo> #<syntax:1:17 1> . #<syntax (2 3 4 5)>)
21:14 freakazoid: rudybot: (syntax-e #`(foo #,1 #,@(list 2 3)))
21:14 rudybot: freakazoid: ; Value: (#<syntax:1:13 foo> #<syntax 1> . #<syntax (2 3)>)
21:14 dingfeng: don't u think syntax unquote splicing should be consistent with the stx argument of (define-syntax foo (lambda (stx) ... ?
21:17 freakazoid: hmm
21:18 (quit) jeapostrophe: Ping timeout: 260 seconds
21:19 freakazoid: well, that's not actually a list
21:20 freakazoid: it's nested syntax pairs
21:20 dingfeng: it's a (cons #'1 #'(2 3))
21:21 freakazoid: I was referring to the one with the foo
21:21 freakazoid: you probably want syntax->list instead of syntax-e to flatten syntax pairs
21:22 freakazoid: rudybot: (syntax->list #`(foo 1 #(,@(list 2 3))))
21:22 rudybot: freakazoid: ; Value: (#<syntax:1:17 foo> #<syntax:1:21 1> #<syntax:1:23 #((unquote-splicing (list 2 3)))>)
21:22 freakazoid: err
21:22 freakazoid: too many parens
21:22 freakazoid: rudybot: (syntax->list #`(foo 1 #,@(list 2 3)))
21:22 rudybot: freakazoid: ; Value: (#<syntax:1:17 foo> #<syntax:1:21 1> #<syntax 2> #<syntax 3>)
21:23 dingfeng: u're right that there's negligible practical impact once you know about it and use syntax->list. it's just... not consistent. i'd sortof expect #` to be the same as the stx argument for a macro
21:23 dingfeng: to produce something identical to the stx argument for a macro
21:23 freakazoid: "A syntax object that is the result of read-syntax reflects the use of delimited . in the input by creating a syntax object for every pair of parentheses in the source, and by creating a pair-valued syntax object only for parentheses in the source. See Reading Pairs and Lists for more information."
21:24 freakazoid: I don't know whether/how that applies in this situation - syntax objects are still new to me
21:24 freakazoid: I suspect it's communicating something about the structure of the original code by maintaining the non-flat syntax pair structure
21:26 freakazoid: anyway, time ot head home
21:26 dingfeng: if that's the case, then shouldn't (begin (define-syntax foo (lambda (stx) (print (syntax-e stx)) #'(display "\n"))) (foo 1 2 3)) produce the syntax pairs as well?
21:26 dingfeng: alright. seeya (:
21:27 (quit) dvanhorn: Quit: dvanhorn
21:29 freakazoid: dingfeng: My suspicion about that is that unsyntax-splicing behaves differently than you think it does.
21:29 freakazoid: (whereas in my case, I think it works by magic)
21:32 (quit) freakazoid: Quit: Computer has gone to sleep.
21:35 (quit) mithos28: Quit: mithos28
21:38 (quit) dingfeng: Ping timeout: 265 seconds
22:09 (join) jrslepak
22:15 (join) mithos28
22:16 (quit) mithos28: Client Quit
22:25 (join) jeapostrophe
22:30 (quit) jeapostrophe: Ping timeout: 248 seconds
22:32 (join) freakazoid
22:41 (join) attila_lendvai
22:41 (quit) attila_lendvai: Changing host
22:41 (join) attila_lendvai
23:11 (join) mithos28
23:12 (join) mithos28_
23:13 (join) jeapostrophe
23:15 (quit) mithos28: Ping timeout: 255 seconds
23:15 (nick) mithos28_ -> mithos28
23:20 freakazoid: is there any management of thread mailbox sizes?
23:20 freakazoid: Or is it the same as erlang where you can run yourself out of RAM?
23:21 mithos28: I don't think so, but there are async-channels?
23:21 mithos28: You can use those (ignore the ? from before)
23:21 freakazoid: yeah, I was just wondering if the thread mailbox was the right thing to use for this application
23:22 mithos28: async-channels are built onto of thread queues I think
23:22 (join) dnolen
23:23 freakazoid: hmm
23:24 mithos28: never mind, looked at the code it is built ontop of channels and a manager thread
23:24 freakazoid: yeah, I am looking at that
23:25 mithos28: why do you want thread mailboxes over async-channels?
23:28 (join) jonrafkind
23:31 (quit) jeapostrophe: Ping timeout: 260 seconds
23:36 freakazoid: just seemed like the thread mailbox was more appropriate for what I'm doing
23:37 freakazoid: but an async channel makes more sense to control the mailbox size to the worker threads
23:42 (join) realitygrill
23:54 (join) jeapostrophe
23:55 (quit) dnolen: Quit: dnolen