00:03 (join) mithos28_
00:07 (quit) mithos28: Ping timeout: 240 seconds
00:07 (nick) mithos28_ -> mithos28
00:10 (join) samth_
00:30 (join) dyoo
00:30 (quit) rekahsoft: Ping timeout: 264 seconds
00:31 (join) rekahsoft
00:36 (quit) samth_: Ping timeout: 255 seconds
00:43 (quit) dyoo: Ping timeout: 252 seconds
00:44 (quit) karljoh: Ping timeout: 240 seconds
00:44 (join) dnolen
00:44 (quit) mithos28: Quit: mithos28
00:44 (join) karljoh
00:55 (join) mithos28_
00:59 (quit) mithos28_: Client Quit
01:02 (quit) dnolen: Quit: dnolen
01:04 (join) ckrailo
02:15 (quit) ckrailo: Quit: Leaving...
02:57 (join) hkBst
03:33 (quit) ehine1: 
03:35 (join) noelwelsh
03:56 (quit) eli: Ping timeout: 252 seconds
04:14 (join) eli
04:15 (quit) rekahsoft: Ping timeout: 258 seconds
04:15 (join) rekahsoft
04:26 (join) masm
04:34 ohwow: wow, racket gui is really cool
04:37 (nick) hyko- -> hyko
04:52 ohwow: Is it possible to cross-compile racket programms?
04:52 ohwow: eg: produce a Mac executable on Windows?
05:23 (join) Utkarsh
05:27 (join) MayDaniel
05:37 noelwelsh: I don't think that's possible
05:39 bremner: ohwow: most of the executable is a copy of the racket VM/Jit aiui. You can certainly ship bytecompiled code that runs on all three, if the racket environment is available.
06:08 (join) lucian
06:30 (quit) rekahsoft: Ping timeout: 260 seconds
06:32 (join) rekahsoft
06:34 (quit) MayDaniel: Read error: Connection reset by peer
07:03 eli: ohwow: You can make installers and executables, but each is limited to its own platform.  In the executable case, it's roughly a copy of the racket executable with all the needed materials added to it, so it's impossible to do that (the windows installation doesn't have a mac executable).
07:03 eli: In the installer case, it's not possible because the installer needs the executable too, but also in some cases -- like the osx case -- it needs an OS tool to create that installer (in the osx case it creates a dmg file).
07:08 (nick) _p4bl0` -> _p4bl0
07:08 (quit) em: Ping timeout: 260 seconds
07:11 (join) em
07:13 ohwow: oh
07:13 ohwow: i see
07:13 ohwow: thanks
07:29 (join) jeapostrophe
07:37 Utkarsh: how can I get the first character of a string "foo" ? I tried (first "foo"), but that obviously doesn't work.
07:38 ohwow: rudybot: eval (string-ref "foo" 0)
07:38 rudybot: ohwow: your sandbox is ready
07:38 rudybot: ohwow: ; Value: #\f
07:38 ohwow: rudybot: eval (string-ref "foo" 1)
07:38 rudybot: ohwow: ; Value: #\o
07:38 ohwow: rudybot: eval (string-ref "foo" 2)
07:38 rudybot: ohwow: ; Value: #\o
07:38 ohwow: Utkarsh: ^
07:38 Utkarsh: ohwow: thanks!
07:38 ohwow: np :)
07:51 (quit) rekahsoft: Ping timeout: 252 seconds
07:52 (join) rekahsoft
07:54 (join) anRch
08:07 (join) samth_
08:16 (quit) noelwelsh: Read error: No route to host
08:22 (join) twem2
08:58 (quit) anRch: Quit: anRch
09:20 (quit) jeapostrophe: Quit: jeapostrophe
09:44 offby1: rudybot: doc string-ref
09:44 rudybot: *offby1: your sandbox is ready
09:44 rudybot: *offby1: http://docs.plt-scheme.org/reference/strings.html#(def._((quote._~23~25kernel)._string-ref))
09:45 offby1: Utkarsh: also see ^^
09:50 eli: offby1: "for blah doc blah"
09:58 offby1: rudybot: for eli doc cons
09:58 rudybot: eli: your sandbox is ready
09:58 rudybot: eli: http://docs.plt-scheme.org/reference/pairs.html#(def._((quote._~23~25kernel)._cons))
09:58 offby1 nods sagely
09:59 offby1: hmm
09:59 offby1: rudybot: (define x 3)
09:59 rudybot: *offby1: your sandbox is ready
09:59 rudybot: *offby1: Done.
09:59 offby1: rudybot: x
09:59 rudybot: *offby1: ; Value: 3
09:59 offby1: rudybot: for snord doc ford
09:59 rudybot: snord: your sandbox is ready
09:59 rudybot: snord: not found in any library's documentation: ford
09:59 offby1: rudybot: for horde doc ford
09:59 rudybot: horde: your sandbox is ready
09:59 rudybot: horde: not found in any library's documentation: ford
09:59 offby1: rudybot: for splode doc ford
09:59 rudybot: splode: your sandbox is ready
09:59 rudybot: splode: not found in any library's documentation: ford
09:59 offby1: rudybot: x
09:59 rudybot: *offby1: ; Value: 3
09:59 offby1: hmm
10:00 offby1: I thought I had a small limit on the number of sandboxes, and that "for X doc Y" would essentially DOS that table
10:00 offby1: by creating new sandboxes it'd evict the older ones.
10:02 offby1: I guess we "date" sandboxes by when they're last used, not when they're created.
10:41 (quit) samth_: Ping timeout: 255 seconds
11:00 danking: So, I came across this quote in HtDP's preface: "Second, good text-oriented input requires deep programming skills, which are best acquired in a course on computational problem-solving. Teaching bad text-oriented input is a waste of the teachers' and the students' time."
11:01 danking: It's not immediately obvious what "deep programming skills" text-oriented input requires, could some elucidate what Matthias, et al. is getting at?
11:03 offby1 glances around nervously.
11:07 danking: Shame matthias doesn't frequent #racket.
11:09 bremner: danking: "good text-oriented input" means something like "readline", to me.
11:09 (quit) hkBst: Remote host closed the connection
11:11 bremner: maybe it's a reference to error handling
11:12 bremner: ok, enough wild speculation from me.
11:13 offby1: bremner: I was going to jokingly say that it meant "reimplementing emacs"
11:14 bremner: well, pretty much. Except I wasn't joking.
11:15 bremner: of course it isn't clear that DrRacket meets these standards ;)
11:15 (join) mithos28
11:15 (join) ckrailo
11:16 danking: The doctor bothers me, on the one hand I like the run/interact approach better than standard repls, but I always long for my Emacs modes.
11:16 bremner: danking: geiser has run-interact
11:16 offby1: yeah, geiser is pretty slick
11:20 (join) racket
11:20 racket: [racket] plt pushed 4 new commits to master: http://bit.ly/lYhi5O
11:20 racket: [racket/master] fix `colorize' nested inside `(linewidth #f ...)' - Matthew Flatt
11:20 racket: [racket/master] add mode to `arc' in `dc-path%' to support right/bottom alignment - Matthew Flatt
11:20 racket: [racket/master] fix interaction of `port-commit-peeked' and port positions... - Matthew Flatt
11:20 (part) racket
11:20 danking: bremner: I am thoroughly intrigued.
11:27 penryu: geiser is a quack alternative?
11:30 penryu reads
11:31 (join) mithos28_
11:33 stamourv: penryu: I believe they can be used together, but I haven't tried.
11:34 (join) dnolen
11:35 (quit) mithos28: Ping timeout: 250 seconds
11:35 (nick) mithos28_ -> mithos28
11:38 (nick) samth_away -> samth
11:39 offby1: penryu: it's the first interactive Emacs mode for scheme that I really liked.
11:39 offby1: it's not quite slime, but it's got some nice features
11:39 samth: offby1: it's not the best name, i agree
11:40 samth: but you should mostly use it (match-lambda**) over match-lambda/match-lambda*
11:40 offby1: huh, I've used match-lambda a couple times.  Don't think I've ever used match-lambda*
11:40 offby1: And I will bet a silk pajama there isn't any three-star match-lambda***
11:41 samth: it's for multiple arguments
11:41 samth: danking: i think his point is that parsing/processing text is a hard problem, not a trivial one
11:42 samth: and that a "user interface" that consists of dumb string comparisons is a waste of everyone's time
11:51 Demosthenes: speaking of user interfaces, anyone using racket/gui for gui forms / data entry?
11:52 (quit) mithos28: Quit: mithos28
11:56 offby1: //me watches the tumbleweeds skitter down Main St
11:58 samth: Demosthenes: I remember hearing that Doug Williams was doing something like that
11:59 samth: but i think it was an internal app
12:00 (join) dyoo
12:01 dyoo: I'd like to test a small patch I'm making to the Scribble reader https://github.com/dyoo/racket/commit/1d99e96d2e1809850b2f86ba8cbe3242bb5e6bc4
12:01 rudybot: http://tinyurl.com/3pzzvn8
12:01 dyoo: what test file can I use to make sure I haven't broken anything?
12:03 Demosthenes: samth: yeah, it'd be interenal here... i don't do much gui crud, i prefer text ;] but i'm trying to facilitate text data entry for an end user, and rather than go relearn a compiled language or perl/tk again, i thought racket might work
12:07 samth: dyoo: try running 'raco setup scribblings'
12:08 samth: also, see collects/tests/scribble
12:09 samth: dyoo: in particular, reader.rkt in that directory
12:09 dyoo: thanks
12:09 dyoo: will do; waiting for Racket to finish compiling... :)
12:22 (quit) gienah: Quit: leaving
12:30 (join) MayDaniel
12:33 (quit) dyoo: Quit: Page closed
12:49 (quit) MayDaniel: Read error: Connection reset by peer
13:16 (join) mithos28
13:17 (quit) rekahsoft: Ping timeout: 244 seconds
13:19 (join) rekahsoft
13:19 danking: samth: Hmm, that makes sense. I suppose the text input is really a language and desrves all the careful planning that we put into one.  Also, thanks for the essays, I read the Typed Racket one and I've been slogging through Zigguarat.
13:25 (quit) dnolen: Ping timeout: 252 seconds
13:36 eli: offby1: no DOS -- it only allows you to do it.
13:36 eli: danking: Another point is that when it gets to text, people quickly jump to side effects.
14:05 (quit) mithos28: Quit: mithos28
14:09 (join) racket
14:09 racket: [racket] plt pushed 1 new commit to master: http://bit.ly/kDNFcP
14:09 racket: [racket/master] added example of making an evaluator for interaction - Danny Yoo
14:09 (part) racket
14:10 (quit) DT``: Ping timeout: 255 seconds
14:13 (join) dnolen
14:14 (quit) lucian: Remote host closed the connection
14:15 (join) lucian
14:16 (join) dnolen_
14:20 (quit) dnolen: Ping timeout: 252 seconds
14:25 (join) racket
14:25 racket: [racket] plt pushed 1 new commit to master: http://bit.ly/kTLSai
14:25 racket: [racket/master] Fix #%app bindings for `with-contract'-based contract forms.... - Stevie Strickland
14:25 (part) racket
14:26 (join) mithos28
14:30 (join) mceier
14:36 (join) racket
14:36 racket: [racket] plt pushed 1 new commit to master: http://bit.ly/kHC3Gm
14:36 racket: [racket/master] `procedure-arity-includes?' reports #f for keyword-requiring procs... - Matthew Flatt
14:36 (part) racket
14:42 (quit) mithos28: Quit: mithos28
14:49 (join) mithos28
14:49 (quit) fmu: Quit: Coyote finally caught me
14:55 (join) DT``
15:04 (join) fmu
15:05 (join) tauntaun
15:13 (quit) lucian: Quit: Ex-Chat
15:13 (join) lucian
15:15 (quit) Demosthenes: Quit: leaving
15:17 (join) Demosthenes
15:37 (quit) mithos28: Quit: mithos28
15:38 (join) mithos28
15:56 (join) jao
16:34 (quit) mithos28: Quit: mithos28
16:37 (join) anRch
16:42 (join) mithos28
16:52 (join) racket
16:52 racket: [racket] plt pushed 1 new commit to master: http://bit.ly/kXGYqx
16:52 racket: [racket/master] Fix names in test to make them test-specific. - Stevie Strickland
16:52 (part) racket
16:56 mithos28: stamourv or samth: ping
16:56 stamourv: mithos28: pong
16:56 mithos28: I am getting weird behavior when adding a tests for 'false?'
16:56 stamourv: Such as?
16:57 mithos28: I give both false? and not the same type
16:57 stamourv: Which makes sense.
16:57 mithos28: and they return different types when applied to #t
16:57 stamourv: Odd. Could you lisppaste an example?
16:57 mithos28: (if (false? #t) 1 2)
16:58 mithos28: with 5.1.1 that should trigger the begin bug with if
16:58 mithos28: it doesn't
16:58 mithos28: because the type is (tc-result Boolean (Top | Top) -) instead of  (tc-result Boolean (Bot | Top)
16:59 mithos28: not gets it correct
16:59 mithos28: by the type I mean the type of the expression (false? #t)
17:00 stamourv: the begin bug with if?
17:00 mithos28: at the module level printing the test value
17:00 stamourv: Oh, that.
17:01 stamourv: That got fixed.
17:01 stamourv: But you mean that the dead code optimization is triggered with `not' but not with `false?'
17:01 stamourv: >
17:01 stamourv: s/>/?/
17:03 mithos28: I cannot tell because the version with my changes has the dead code bug fixed
17:03 mithos28: but the false bug is observable in 5.1.1 because the dead code bug does not trigger
17:04 stamourv: Here's a trick, run your code with --log-optimizations as an option to the program. The optimizer should log all the optimizations it does.
17:04 stamourv: That way you should know whether the optimization happens or not.
17:05 stamourv: (As of current master, that's a debugging feature, but I'm currently turning it (and more) into a user-usable feature. I hope to land that branch today or tomorrow.)
17:06 mithos28: http://pastebin.com/TXcgdBGm
17:06 (join) racket
17:06 racket: [racket] plt pushed 1 new commit to master: http://bit.ly/jinkJo
17:06 racket: [racket/master] Scribble Latex Unicode fixes - Matthew Flatt
17:06 (part) racket
17:07 mithos28: Here is a program and the output with log-optimizations
17:07 mithos28: it optimizes for not and doesn't for false?
17:07 stamourv: Ok, the contradiction is only detected for `not'.
17:08 stamourv: Let me check the code for contradiction detection. Maybe it special cases not.
17:08 mithos28: the problem is with false?
17:09 mithos28: because of how I found it, i can see that it doesn't know that false? cannot return a false value when applied to #t
17:10 stamourv: Wait, so what type did you give to each of them?
17:11 mithos28: the unit test (tc-e (false? #t) #:ret (ret B (-FS -bot -top))) fails
17:11 mithos28: sorry I mean cannot return a true value
17:11 mithos28: (make-pred-ty B)
17:11 mithos28: (make-pred-ty B ) (no smilies)
17:12 stamourv: That's wrong.
17:12 mithos28: ugh
17:12 stamourv: That's the type of `boolean?'.
17:12 mithos28: sorry I cannot type
17:12 mithos28: or think
17:12 mithos28: it is actually (make-pred-ty (-val #f))
17:13 stamourv: That sounds reasonable. Is that what you're using?
17:13 mithos28: yes
17:13 mithos28: that was the type for false?
17:13 mithos28: I changed not to that type, and added tests. false? failed them and not passed them
17:14 stamourv: I think I found the problem.
17:14 stamourv: Line 515 of typecheck/tc-app.rkt
17:15 stamourv: `not' is special-cased to flip filters. I'll add `false?' to that.
17:16 stamourv: That should fix it.
17:17 mithos28: is that just changing the not to (~or not false?) so that I can test it
17:17 stamourv: Yes, as long as `false?' is in scope at the right phase.
17:18 stamourv: And it looks like it's the case.
17:18 stamourv: Give it a try.
17:18 mithos28: lots of tests failed
17:20 stamourv: Unit tests?
17:20 mithos28: yep
17:20 mithos28: 114 fails 16 errors
17:20 stamourv: Ok. I'll finish what I'm doing, and then I'll try it on my end.
17:21 mithos28: k
17:26 mithos28: it looks like false? needed to be added to the literals
17:27 stamourv: Oh, that makes sense.
17:27 stamourv: Does it work now?
17:27 mithos28: no
17:27 mithos28: it doesn't make all the tests fail
17:27 mithos28: but my test still does
17:28 stamourv: Odd. I'll try it on my end.
17:30 stamourv: Works for me. Both `not' and `false?' trigger the optimization.
17:31 stamourv: However, I don't have your patches for the new types.
17:32 stamourv: Now trying with both having the `make-pred-ty' type.
17:32 stamourv: Works fine too.
17:32 stamourv: Did you rebuild?
17:33 mithos28: typed/racket and typed-scheme
17:33 stamourv: Hmm, that should work.
17:33 mithos28: are you on master?
17:33 mithos28: I'm a couple days old
17:33 stamourv: Master as of earlier today.
17:34 stamourv: Anyway, I've got a working patch here, so I'll push that, and hopefully that will fix it on your end too.
17:34 mithos28: ok
17:34 mithos28: if not I'll get in touch again
17:34 mithos28: thanks for the help
17:37 stamourv: No problem.
17:37 stamourv: I'm running the tests. If it all works, I'll push.
17:38 (quit) mithos28: Quit: mithos28
17:46 (quit) anRch: Quit: anRch
18:01 (quit) dnolen_: Ping timeout: 252 seconds
18:09 (join) racket
18:09 racket: [racket] plt pushed 1 new commit to master: http://bit.ly/kF1jcL
18:09 racket: [racket/master] Unify types and filter behavior of not and false?. - Vincent St-Amour
18:09 (part) racket
18:18 (quit) lucian: Remote host closed the connection
18:48 (join) dnolen
18:50 (join) dyoo
18:51 (join) mithos28
18:53 (quit) ckrailo: Quit: Leaving...
18:54 (join) ckrailo
19:14 (quit) dyoo: Ping timeout: 252 seconds
19:22 (quit) masm: Quit: Leaving.
19:23 (quit) tauntaun: Ping timeout: 255 seconds
19:44 (join) gienah
19:45 (nick) samth -> samth_away
19:55 (join) tauntaun
20:08 (quit) dnolen: Quit: dnolen
20:14 (quit) ckrailo: Quit: Leaving...
20:24 (quit) mceier: Quit: leaving
20:35 (join) racket
20:35 racket: [racket] plt pushed 1 new commit to master: http://bit.ly/k5yxg6
20:35 racket: [racket/master] fix JITted multi-valued `begin0' in a single-value context... - Matthew Flatt
20:35 (part) racket
20:54 (join) samth_
21:18 (quit) rekahsoft: Read error: Operation timed out
21:18 (join) rekahsoft
21:38 (join) racket
21:38 racket: [racket] plt pushed 7 new commits to master: http://bit.ly/l2KFtz
21:38 racket: [racket/master] Add activity graphs from gmane. - Eli Barzilay
21:38 racket: [racket/master] Add an error message when the sandbox doesn't capture its output as a string. - Eli Barzilay
21:38 racket: [racket/master] Switch to `#lang', reformat, and some minor improvements. - Eli Barzilay
21:38 (part) racket
21:42 (join) hoeppner
21:52 hoeppner: Is there a place chat logs are stored?
21:53 (quit) tauntaun: Quit: Ex-Chat
21:53 DT``: hoeppner, http://racket-lang.org/irc-logs/ ?
21:53 hoeppner: nvm looked
21:54 hoeppner: oh thanks
21:54 hoeppner: in the title :S
21:54 DT``: oh, right.
21:54 DT``: I didn't look at the topic either. lol
22:03 (quit) hoeppner: Ping timeout: 255 seconds
22:18 (join) neilv
22:41 (join) no-name-
22:41 (part) no-name-: "Leaving"
22:43 (join) no-name-
22:56 (quit) samth_: Ping timeout: 255 seconds
22:57 (quit) DT``: Ping timeout: 255 seconds
22:58 (quit) gienah: Quit: leaving
23:11 (join) DT``
23:38 (quit) no-name-: Remote host closed the connection
23:47 (quit) rekahsoft: Ping timeout: 240 seconds
23:48 (quit) mithos28: Quit: mithos28
23:49 (join) rekahsoft
23:50 (join) tauntaun