00:12 (join) leo2007
00:49 (join) yoklov
00:50 (join) yoklov1
00:50 (quit) yoklov: Read error: Connection reset by peer
01:04 (quit) yoklov1: Quit: Leaving.
01:21 (quit) dnolen: Quit: dnolen
01:30 (join) Demosthenes
01:30 (quit) realitygrill: Quit: realitygrill
02:08 (quit) jonrafkind: Read error: Operation timed out
02:11 (quit) Demosthenes: Ping timeout: 260 seconds
02:32 (join) qha
02:35 (quit) leo2007: Quit: rcirc on GNU Emacs 23.3.50.1
03:05 (join) hkBst
03:07 (quit) hkBst: Read error: Connection reset by peer
03:09 (join) hkBst
03:09 (quit) hkBst: Changing host
03:09 (join) hkBst
03:47 (quit) blomqvist: Read error: Operation timed out
03:47 (join) blomqvist
03:49 (quit) chemuduguntar: Read error: Connection reset by peer
04:23 (join) ravi_
04:32 (quit) ckrailo: Quit: Computer has gone to sleep.
05:23 ravi_: hello
05:23 ohwow: hi
05:24 ravi_: hi ohwow
05:57 (join) masm
07:03 (quit) Lectus: Quit: Ex-Chat
07:37 (join) lucian
07:47 (join) tildedave
07:50 (quit) lucian: Read error: Operation timed out
08:05 (join) lucian
08:10 (join) Lajla
08:10 (part) Lajla
08:26 (quit) lucian: Remote host closed the connection
08:28 (join) wesh
08:28 wesh: cc bb
08:28 wesh: hi
08:29 wesh: someone can tell me where i can found chipmunk demo's plz? ^^"
08:46 wesh: bande de tepu
08:46 wesh: rep plz!
08:50 (join) leo2007
09:12 (quit) wesh: Ping timeout: 252 seconds
09:26 (join) mithos28
09:27 (join) dnolen
09:27 (join) PLT_Notify
09:27 PLT_Notify: racket: master Matthew Flatt * a60bfe9 (2 files in 2 dirs): remove some obsolete Mac OS X compiler flags for gracket
09:27 PLT_Notify: racket: master Matthew Flatt * 5b2492e (1 files in 1 dirs): places, creating a subprocess: reset SIGCHLD correctly
09:27 PLT_Notify: racket: master commits f60f234...5b2492e - http://bit.ly/kZsBNL
09:27 (part) PLT_Notify
09:30 (quit) dnolen: Remote host closed the connection
09:39 (join) dnolen
09:40 (quit) mithos28: Quit: mithos28
09:49 (join) PLT_Notify
09:49 PLT_Notify: racket: master Robby Findler * b6fc13b (1 files in 1 dirs): a start on the help docs
09:49 PLT_Notify: racket: master Robby Findler * e41deb5 (1 files in 1 dirs): more help collection documentation
09:49 PLT_Notify: racket: master Robby Findler * 553d9ff (2 files in 2 dirs): made set/c preserve flat-contract'ness of its argument
09:49 PLT_Notify: racket: master commits 5b2492e...553d9ff - http://bit.ly/m9s0yA
09:49 (part) PLT_Notify
09:57 (join) yoklov
10:20 (nick) samth_away -> samth
10:53 (quit) hkBst: Read error: Connection reset by peer
10:54 (quit) ravi_: Ping timeout: 260 seconds
11:08 (join) ravi_
11:13 (join) evhan_
11:13 (join) realitygrill
11:15 (nick) evhan_ -> evhan`
11:19 (join) ckrailo
11:19 (quit) evhan`: Quit: leaving
11:45 (join) PLT_Notify
11:45 PLT_Notify: racket: master Matthew Flatt * a4da2a3 (8 files in 3 dirs): fix `varref' in `compiler/zo-structs', etc. ... - http://bit.ly/jzEZaW
11:45 (part) PLT_Notify
11:46 (join) tauntaun
11:54 (quit) leo2007: Ping timeout: 260 seconds
11:55 (join) vu3rdd
11:55 (quit) vu3rdd: Changing host
11:55 (join) vu3rdd
12:14 (join) jonrafkind
12:21 (quit) yoklov: Quit: Leaving.
12:26 (quit) tauntaun: Quit: Ex-Chat
12:27 (join) leo2007
12:43 (join) tauntaun
12:58 (quit) martinhex: Read error: Connection reset by peer
13:01 (quit) chturne: Quit: ChatZilla 0.9.86.1 [Firefox 4.0.1/20110413222027]
13:04 (join) martinhex
13:44 (join) anRch
13:58 (quit) tauntaun: Ping timeout: 240 seconds
14:19 (quit) leo2007: Quit: rcirc on GNU Emacs 23.3.50.1
14:24 (join) PLT_Notify
14:24 PLT_Notify: racket: master Matthew Flatt * 8ad8d5b (2 files in 1 dirs): fix tl-map parse & marshal - http://bit.ly/iYspXB
14:24 (part) PLT_Notify
14:34 (quit) anRch: Quit: anRch
14:36 (quit) vu3rdd: Read error: Connection reset by peer
14:37 (join) vu3rdd
14:37 (quit) vu3rdd: Changing host
14:37 (join) vu3rdd
14:53 (join) lucian
15:07 (quit) ravi_: Ping timeout: 258 seconds
15:11 (join) dzhus
15:16 (quit) dzhus: Remote host closed the connection
15:21 (join) yoklov
15:22 (join) yoklov1
15:22 (quit) yoklov: Read error: Connection reset by peer
15:22 (quit) yoklov1: Client Quit
15:41 (join) ravic
15:41 (nick) ravic -> chemuduguntar
15:58 (join) mithos28
16:06 (join) Nican
16:18 (join) devogon
16:21 (join) flo1991
16:22 flo1991: hi gus, i need some help!
16:22 samth: hi flo1991, just ask away
16:22 flo1991: i just updated my dr.racket , foolish as i am i forget to back up a project that i was working on which was saved in the collects folder
16:23 flo1991: does anyone know how to retrieve my previous collects folder?
16:23 samth: did you delete the folder?
16:23 samth: the old drracket install, that is
16:23 flo1991: i guess the dr.racket installer did that for me since i can't find it
16:24 samth: where is drracket installed on your computer?
16:25 flo1991: c:\program files
16:26 samth: if you look in that directory, do you see only one version of Racket?
16:26 flo1991: yes
16:30 (join) PLT_Notify
16:30 PLT_Notify: racket: master Matthew Flatt * ca16f25 (1 files in 1 dirs): fix `in-list' to check argument with `list?' ... - http://bit.ly/iNtohw
16:30 (part) PLT_Notify
16:34 (quit) flo1991: Ping timeout: 252 seconds
16:34 (join) PLT_Notify
16:34 PLT_Notify: racket: master Robby Findler * 0e58a37 (1 files in 1 dirs): fixed up a test (that I hadn't meant to push the last time, sigh) - http://bit.ly/mDi40u
16:34 (part) PLT_Notify
16:37 (quit) mithos28: Quit: mithos28
17:18 (quit) dnolen: Quit: dnolen
17:26 (quit) jonrafkind: Ping timeout: 276 seconds
17:40 (join) mithos28
17:59 (join) PLT_Notify
17:59 PLT_Notify: racket: master Matthias Felleisen * b74c9f9 (3 files in 2 dirs): ensure that dimensions of canvas are below 2000 by 2000; closes PR 11907 - http://bit.ly/iDo1lV
17:59 (part) PLT_Notify
18:03 (quit) mithos28: Quit: mithos28
18:06 (quit) qha: Quit: Leaving.
18:11 (join) fmu`
18:11 (quit) fmu: Ping timeout: 246 seconds
18:12 rotty: eli: how does one generate the "textual" stripped-down source tarball from a git checkout?
18:21 (join) qha
18:26 (quit) martinhex: Read error: Connection reset by peer
18:27 (join) martinhex
18:28 (part) lnostdal: "Leaving"
18:29 (quit) lucian: Remote host closed the connection
18:33 (join) dnolen
18:50 eli: rotty: So, what do you need it for?
18:50 rotty: eli: I want to create a stripped-down installation for a shared server running Debian Lenny.
18:51 eli: Is it only for server things?
18:51 rotty: yes.
18:51 rotty: no display attached to that machine, and no need at all for running X11 client stuff.
18:51 eli: Did you start from the textual distribution?
18:52 rotty: no. I could do that, but I'd like to be able to recrate the textual distribution myself from master, in case I need to tweak something in the R6RS area.
18:53 rotty: s/recrate/recreate/
18:53 eli: Is it practical to use the nightly builds?
18:54 rotty: I'd really prefer to start from my local git repo.
18:55 eli: If you want a git repo then you're already talking about a pretty big size.
18:55 eli: For example, including all the gui stuff and all the doc sources.
18:55 eli: (And tests, and games, etc.)
18:56 rotty: I build and deploy on different machines
18:56 rotty: for the build machine, size is irrelevant; for the deployment machine, not so much.
18:57 (join) mithos28
18:58 (quit) mithos28: Client Quit
18:58 eli: In that case, I think that the best thing to do is to grab a nighly build distribution, see the files in it, and make a script that removes the files that are not there but in the plt repo.
18:59 rotty: (what I've done now is to build and install (using DESTDIR) on the build machine, strip the worst size-wise offenders from collects/ manually, pack that up, transfer to the target machine, and unpack there)
18:59 eli: The thing is that the script that creates these things is pretty difficult to run.  (You can find it in collects/meta/dist*)
19:00 rotty: yeah, I've peeked around there, but couldn't figure out how to actually use this stuff, hence the original question)
19:00 eli: It's *much* easier to avoid DESTDIR hacks in instead do the default in-place build.
19:00 rotty: well, it worked all very nicely, as far as i can tell
19:00 eli sighs
19:00 rotty: (and DESTDIR is not a hack, IMHO)
19:01 eli: The in-place build is an extremely convenient way to do these things, yet people stick with DESTDIR and --prefix and such...  I really can't figure out why.
19:02 eli: In any case, that's why I said that just doing that subtraction using the textual distribution only as a file list is easier.
19:02 rotty: as mentioned, I don't want to build on the deployment machine.
19:02 rotty: yeah, that's a good way, but quite a hack.
19:03 eli: Huh?
19:03 rotty: s/good/workable/
19:03 (quit) rudybot: Ping timeout: 260 seconds
19:03 (quit) offby1: Ping timeout: 276 seconds
19:03 eli: You don't build on the deployment machine.
19:03 eli: You build a tree, then do whatever you want with it -- and the fact that it's all in a single directory makes it far more convenient than the whole FHS mess.
19:03 rotty: well, relying on the nightly builds as a filelist provider is a hack, IMO.
19:04 rotty: but can I just transfer that tree to a different machine and use it in a different directory?
19:04 eli: I was referring to the "don't want to build on the deployment machine".  The distribution thing that I talked about is a hack, yes.
19:05 eli: Yes, absolutely.  That's exactly why it's so convenient.
19:07 rotty: ok, good to know that Racket doesn't embed any absolute file names in build products of an in-tree build.
19:08 eli: Yeah -- and very intentionally so.
19:08 (join) mithos28
19:09 eli: If you do a grep you will find a few strings, but those are mostly results of syntax expansions -- and should be abolished eventually too.
19:09 (quit) mithos28: Client Quit
19:09 eli: In fact, the installers that we do are made from in-place builds.
19:10 rotty: thanks for the advice, eli.
19:11 (nick) fmu` -> fmu
19:13 rotty: nevertheless, I look forward to having an easier way to obtain a "core" (textual) installation from a git checkout (as I've heard, there are plans to modularize Racket in such a way)
19:21 eli: rotty: Yes, there are.
19:43 (quit) Eren: Ping timeout: 258 seconds
19:46 (quit) devogon: Remote host closed the connection
20:14 (join) rpr`
20:15 rpr`: samth: Have you guys even started to think about bring in classes/objects into typed land?
20:18 samth: rpr`, yes, we have
20:18 samth: there's a little experimental support for talking about exisisting classes, and instantiating/sending to them
20:18 samth: and there's a grad student here at northeastern who's working on the larger problem
20:19 rpr`: What is the larger problem?
20:19 samth: rpr`, see here for the current support: http://pre.plt-scheme.org/docs/html/ts-reference/index.html#%28part._.Experimental_.Features%29
20:20 samth: the larger problem is full integration -- writing typed classes, etc
20:20 rpr`: I could get far with just require/tuped object? in to typed land and with typing message sigs.
20:21 rpr`: Don't need typing class hiers or defining typed methods/attr.  Though would be nice.
20:21 samth: rpr`, in that case you can use the current support
20:21 samth: although the docs are pretty thin
20:22 rpr`: Canonical prototype of need: pgsql module is basically a single class with 6 methods.
20:22 rpr`: Ok let me read first.
20:22 samth: you can see some examples of how it works here: https://github.com/plt/racket/blob/master/collects/drracket/private/auto-language.rkt
20:23 rpr`: Not that its my field or anything, but I'd have bet money what you guys are trying to do well, couldn't be done. i.e. no type theory would get you there.
20:25 rpr`: But it seems to work pretty well, though I haven't tried a byzantine attempt to subvert the type system.  Syntax is clean as well.  Impressive.
20:39 rpr`: I don't use racket's classes.  but spgslq has (define connnection% (class (...) ...) and I modified their source file to provide connection%.  I also a small procedure to (define (quick-connect db user pass) which connects and returns an instance of connection%.
20:41 (quit) vu3rdd: Remote host closed the connection
20:41 rpr`: so (require/typed  (planet ... /spgsql) (quick-connect (String String String -> Instance connection%) should work?
20:42 rpr`: Type Checker: Unbound type name connection% in: connection%
20:46 samth: you're missing some parens
20:47 samth: and you have to define a type called `connection%' for that to work
20:47 rpr`: how do I do that?
20:47 samth: take a look at the example i mentioned
20:47 samth: and at the files it requires
20:47 samth: in particular, typed/mred/mred
20:47 rpr`: ok
20:47 samth: gotta run for a little while
20:48 samth: feel free to send me email
20:48 rpr`: I'll get it.  Thanks.
21:01 rpr`: We got a bingo.
21:01 rpr`: Complicated they used several mixins and the kitchen sink in laying out their connection%.
21:05 (quit) masm: Read error: Operation timed out
21:18 (quit) realitygrill: Ping timeout: 240 seconds
21:20 (join) realitygrill
21:21 (join) tauntaun
21:28 (join) mithos28
21:35 (join) jonrafkind
21:37 (quit) ckrailo: Quit: Computer has gone to sleep.
21:41 (join) grettke
21:43 (quit) tauntaun: Quit: Ex-Chat
21:49 (quit) grettke: 
22:04 rpr`: hmm how does one efficiently or otherwise cast a (Listof Any) to a (Listof String)???
22:06 jonrafkind: map?
22:06 DT``: is it possible?
22:07 jonrafkind: i thought thats what the point of occurence typing was, to convert any into some type simply by using a predicate
22:07 rpr`: I think (assert ...) will do it.
22:16 rpr`: so far nada.
22:18 (quit) mithos28: Quit: mithos28
22:22 (join) realitygrill_
22:23 (quit) realitygrill: Read error: Connection reset by peer
22:23 (nick) realitygrill_ -> realitygrill
22:23 rpr`: can't seem to put together a (Any -> Boolean : (Listof String)) that it likes which I can use with assert.  I'll try the map approach.
22:29 (join) mithos28
22:33 rpr`: map works.
22:34 rpr`: But when you have 116,000 strings from a query that extra map is pretty heavy.
22:34 DT``: rpr`, (define-predicate list-string? (List String)).
22:35 DT``: (assert list list-string?)
22:35 rpr`: did you try that?
22:35 DT``: yeah, it did work.
22:36 rpr`: Darn  I tried some variation of (Any -> Boolean : (Listof String))
22:36 rpr`: Yours is too easy. :(
22:37 DT``: I'm sure that it has to traverse the list anyway.
22:37 DT``: or maybe not.
22:40 samth: rpr`, you can use andmap as well
22:40 samth: or filter
22:42 rpr`: OK  I didn't know why DT``'s idea worked.  list-string? is pre-existing.
22:43 (quit) evhan: Quit: leaving
22:43 DT``: reference to an identifier before its definition: list-string? <--- ?
22:43 (join) evhan
22:45 (join) vu3rdd
22:45 (quit) vu3rdd: Changing host
22:45 (join) vu3rdd
22:45 rpr`: samth: If you can translate this technology into typed javascript in mozilla.  You will be worshipped as a living god for decades.
22:47 rpr`: and for your second miracle, get modules in there as well.
22:48 samth: rpr`, your wish is my command: http://wiki.ecmascript.org/doku.php?id=harmony:modules
22:48 samth: that by my and (mostly by) Dave Herman at Mozilla
22:48 samth: :0
22:48 (nick) samth -> samth_away
22:50 rpr`: was that for some version ecma that never got approved or will we actually see real soon now in FF.
22:51 rpr`: typed javascript would be special.
23:07 rpr`: Did some tests andmap appears to be the way to go when you need a predicate to cast a (Listof a) to a (Listof b).
23:07 rpr`: You get the traversable, but no duplication of the list.
23:36 (join) Demosthenes
23:40 (quit) Demosthenes: Ping timeout: 252 seconds
23:46 (quit) chemuduguntar: Ping timeout: 258 seconds