00:00 lewis1711: rapacity: aha, something like that yeah. also want to disallow defining the variable
00:00 lewis1711: so can add that to the conds
01:00 (join) mheld
01:00 (nick) abbe -> writer
01:00 (quit) writer: Quit: writer
01:36 (join) lewis17111
01:36 (quit) lewis1711: Ping timeout: 265 seconds
01:38 (join) jeapostrophe
01:42 (quit) lewis17111: Ping timeout: 265 seconds
01:45 (quit) mheld: Ping timeout: 255 seconds
01:57 (join) lewis1711
01:58 (join) jonrafkind
02:02 (quit) lewis1711: Ping timeout: 260 seconds
02:12 (quit) offby1: Ping timeout: 240 seconds
02:13 (quit) jeapostrophe: Quit: jeapostrophe
02:17 (join) lewis1711
02:26 (quit) lewis1711: Ping timeout: 260 seconds
02:59 (quit) jonrafkind: Ping timeout: 276 seconds
03:02 (quit) mwolfe: Ping timeout: 255 seconds
05:01 (join) sunnyps
05:20 (join) neilv
05:20 (quit) neilv: Changing host
05:20 (join) neilv
06:28 (join) masm
06:53 (join) Roverd
06:53 Roverd: Has racket the ability to use a a sorting or min/max function together with a lambda. I couldn't find it in the documentation?
07:29 (join) etate
07:30 etate: hey, is it possible to have multiple separate threads run separate instances of racket? In the docs it seems to point to not being able to do this, if so is there a good reason why?
07:36 (quit) sunnyps: Ping timeout: 265 seconds
07:43 (join) MayDaniel
07:50 (quit) Roverd: Quit: Page closed
08:06 bremner: etate: usually people would call those processes rather than threads
08:07 bremner: maybe http://docs.racket-lang.org/more/index.html is helpful?
08:08 (quit) MayDaniel: Read error: Connection reset by peer
08:22 (join) abbe
08:22 (quit) abbe: Changing host
08:22 (join) abbe
08:24 (join) stis
08:25 (join) sunnyps
08:28 etate: bremner: why would they call them processes instead of thread? I'm talking about spawning multiple instances of Racket in separate OS level threads here, from C.
08:29 (join) jeapostrophe
08:36 neilv: are you sure you want to embed racket? afaik, almost everyone right now makes racket the main process and then calls out to c code via the ffi
08:38 etate: neilv: yes, because I want to control, from outside racket, many instances of it (eg to do multicore stuff)
08:39 neilv: you might have to use multiple OS processes, one racket per process, and do ipc
08:39 etate: yeah, thats basically what i'm trying to do, but i want to be able to do it from inside another program
08:40 etate: so instead of using unix, i want to spawn a C thread and make it do basically the same thing
08:40 etate: but according to the docs, you can't call the scheme primitives from multiple OS level thread
08:40 etate: threads*
08:41 neilv: os threads are lighter weight than os processes, but i think you might have better luck just using multiple os processes
08:42 neilv: os processes might be lighter-weight than you think. and, if they are short-lived enough that startup overhead is expensive, you might be able to do pools of reusable processes
08:47 etate: neilv: aaaaah i've been looking at the wrong level of abstraction
08:47 etate: neilv: maybe i _have_ to fork these things instead of thinking about pthreads
08:49 neilv: yeah, you could probably fork with racket embedded, but fork&execing racket and avoiding embedding might be less painful
08:49 etate: I was sure there was a good reason for not being able to instantiate two rackets at the pthread level, maybe this is it
08:50 etate: yeah, i also was thinking that this whole embedding business is OTT
08:50 neilv: i suspect that your development experience of the racket side of the code will be much better if it doesn't require embedding
08:50 neilv: plus you avoid the messy c coding, and having a garbage collector stomping all over your c code
08:50 etate: well the problem is i'm trying to make a generic framework for spawning processes of any kind, not just racket
08:50 etate: thats why the C code is at some points inevitable
08:51 etate: i should say: spawning and communicating
09:00 (join) mceier
09:04 Demosthenes: anyone have an example of call-with-transaction in snooze? i'm trying to speed up a series of inserts
09:12 (join) losvedir
09:37 (quit) jeapostrophe: Quit: jeapostrophe
09:38 (join) jeapostrophe
09:41 (quit) etate: Quit: Leaving
10:04 (quit) jeapostrophe: Quit: jeapostrophe
10:05 (join) jeapostrophe
10:05 (join) anRch
10:06 (join) MayDaniel
10:37 (quit) jeapostrophe: Quit: jeapostrophe
10:42 (join) offby1
10:43 (quit) offby1: Changing host
10:43 (join) offby1
11:05 (quit) MayDaniel: Read error: Connection reset by peer
11:17 (quit) anRch: Quit: anRch
11:53 (join) MayDaniel
12:00 (join) shofetim
12:19 (quit) MayDaniel: 
12:28 (join) mwolfe
12:29 (quit) stis: Ping timeout: 260 seconds
12:50 (join) stis
13:00 (part) neilv: "Leaving"
13:34 (quit) sunnyps: Quit: Leaving
13:35 (join) alexsuraci_
13:36 (nick) alexsuraci_ -> testingfoo
13:38 (quit) testingfoo: Client Quit
14:07 (join) anRch
14:12 (quit) shofetim: Ping timeout: 240 seconds
14:23 (join) shofetim
14:25 (part) shofetim
14:34 (join) jeapostrophe
14:41 (join) MayDaniel
14:57 (quit) anRch: Quit: anRch
15:00 (quit) jeapostrophe: Quit: jeapostrophe
15:05 (join) jeapostrophe
15:07 (quit) MayDaniel: 
15:09 (join) jeapostrophe_
15:09 (quit) jeapostrophe_: Read error: Connection reset by peer
15:10 (join) jeapostrophe_
15:11 (quit) jeapostrophe: Ping timeout: 260 seconds
15:11 (nick) jeapostrophe_ -> jeapostrophe
15:25 (join) jeapostrophe_
15:26 (quit) jeapostrophe: Ping timeout: 272 seconds
15:26 (nick) jeapostrophe_ -> jeapostrophe
15:26 (quit) jeapostrophe: Client Quit
15:32 (quit) parcs: Quit: WeeChat 0.3.3
15:33 (join) parcs
16:08 (join) philo
16:17 (quit) evhan: Read error: Operation timed out
16:19 (quit) mwolfe: Ping timeout: 272 seconds
16:20 (join) evhan
16:23 philo: looking for information on the bytecode format and on the jit
16:24 offby1: spewing random gerunds into the channel
16:29 philo: ?
16:29 philo: the documentions doesnt seems to mention anything
16:37 rapacity: not sure if this is what you want http://plt.eecs.northwestern.edu/racket-machine/
16:38 philo: look good, was that on the site and i missed it ?
16:40 rapacity: I got that link off the mailing list, not sure if its linked somewhere on the site
16:45 (join) palad
16:45 (quit) rudybot: Remote host closed the connection
16:46 (join) rudybot
16:46 palad: hi, im having an issue with let*
16:46 palad: i get bad synthax: (let* ((matches (find-matches operators top-node '() )))            (apply-matches kb stack operators plan matches))
16:51 rapacity: the let* itself looks ok
16:53 (join) philo_
16:55 (quit) philo: Ping timeout: 265 seconds
16:58 danking: samth: Alright, I'll post if I don't forget about it by tonight!
17:00 (quit) tv|z: Ping timeout: 240 seconds
17:01 (quit) losvedir: Quit: losvedir
17:02 (join) tv|z
17:12 (join) MayDaniel
17:17 (join) mwolfe
17:23 (quit) Demosthenes: Ping timeout: 265 seconds
17:25 (join) Demosthenes
18:01 (quit) mceier: Quit: leaving
18:01 (quit) MayDaniel: Read error: Connection reset by peer
18:20 (quit) masm: Quit: Leaving.
18:24 (quit) stis: Remote host closed the connection
18:36 (join) anRch
18:53 (join) losvedir
19:02 Lajla: It's possible to verify for a compiler that funtions which use bangs internally only apply bangs to locally initiated variables and are thus pure, right?
19:09 (quit) anRch: Quit: anRch
19:11 (join) jeapostrophe
19:18 (join) jonrafkind
19:29 (quit) jeapostrophe: Quit: jeapostrophe
19:43 (quit) losvedir: Quit: losvedir
19:43 (quit) parcs: Read error: Connection reset by peer
19:44 (join) parcs
19:52 Lajla: jonrafkind, surely ou know this
19:53 Lajla: <Lajla> It's possible to verify for a compiler that funtions which use bangs internally only apply bangs to locally initiated variables and are thus pure, right?
19:53 jonrafkind: yea
19:54 jonrafkind: well, by pure you mean has no side-affects
19:54 jonrafkind: but you can stil do IO and whatnot
19:54 jonrafkind: but you can definately force functions to only mutate variables introduced in scopes beneath the function
20:10 (join) jeapostrophe
20:19 Lajla: jonrafkind, well, yeah, I mean mutations
20:19 Lajla: I don't mean force, but like say.
20:19 Lajla: Can a compiler prove that even though a function contains some '!' functon, even inside another function that is inside it.
20:20 Lajla: If the mutation performed is local to the function
20:20 Lajla: Like ehh, string and vector I guess in their standard implementation do feature mutation internally.
20:20 jonrafkind: yea its not too hard to do, its called escape analysis in general
20:20 jonrafkind: well ive never implemented it so i dont know how hard it is to do..
20:23 Lajla: But you are jonrafkind, you wield the power cosmic, in all this universe there is nothing you cannot do.
20:23 Lajla: Okay, but next step.
20:23 Lajla: Is it deciable.
20:23 jonrafkind: well, i think so
20:23 Lajla: Like, can a compiler lexically provide a correct yes/no to whether the function can or cannot mutate outside of itself.
20:24 Lajla: =(
20:24 Lajla: Oh
20:24 Lajla: I read 'I don't think so;
20:24 Lajla: Schweet
20:24 jonrafkind: your analysis might have to be conservative at some points
20:25 jonrafkind: because it might not be possible to tell if some variable escapes without executing the program
20:25 jonrafkind: like (let ([b (box (if (omega?) x y))] b)
20:25 jonrafkind: so which escapes, x or y? well its impossible to know, so you have to say both
20:26 jonrafkind: i guess using box is a bad example there
20:26 jonrafkind: well if x and y are created as boxes, that would work I suppose
20:28 (quit) mwolfe: Remote host closed the connection
20:29 Lajla: Hmm
20:29 Lajla: I suppose that falls under 'can'
20:30 Lajla: But yeah, I might need to formalize that better
20:34 (quit) jeapostrophe: Quit: jeapostrophe
20:51 (join) jeapostrophe
20:59 (join) mwolfe
21:37 (quit) mwolfe: Remote host closed the connection
22:10 (quit) philo_: Ping timeout: 265 seconds
22:30 (quit) jeapostrophe: Quit: jeapostrophe
23:55 askhader: So how do you fellow racketeers feel about
23:55 askhader: Ugh enter key
23:55 jonrafkind: 42
23:55 askhader: As I thought.
23:55 askhader: =]
23:58 askhader: The speed of racket =O
23:59 jonrafkind: The fires of london >:E