00:05 (join) jeapostrophe
00:05 (quit) jeapostrophe: Client Quit
00:28 (part) ice_man: "ERC Version 5.3 (IRC client for Emacs)"
00:32 (join) jeapostrophe
00:35 (quit) dnolen: Quit: dnolen
00:37 (quit) jeapostrophe: Client Quit
01:49 (join) dsrguru
01:52 (join) RacketCommitBot
01:52 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://bit.ly/oaAUlV
01:52 RacketCommitBot: [racket/master] updated Lazy Racket to use racket/base - Stephen Chang
01:52 (part) RacketCommitBot
02:08 (join) RacketCommitBot
02:08 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://bit.ly/nKRaYq
02:08 RacketCommitBot: [racket/master] fix lazy.scrbl so it compiles properly - Stephen Chang
02:08 (part) RacketCommitBot
02:27 (quit) dsrguru: Quit: leaving
02:29 (quit) jonrafkind: Ping timeout: 260 seconds
02:43 (join) realitygrill
03:14 (join) hkBst
03:20 (join) lewis1711
04:00 (join) mceier
04:06 (quit) Lajla: Quit: Phallic Illuminatus
04:08 (join) Lajla
04:28 (join) tfb
04:36 (quit) noelw_: Read error: No route to host
04:37 (join) noelw
05:00 (join) masm
05:30 (join) MayDaniel
06:00 lewis1711: https://gist.github.com/1151241 hmm so my vector-foldl does the same thing as sequence-fold which does a different thing from foldl....
06:22 (quit) bmp: Quit: Leaving...
06:25 lewis1711: ohh I had it round the wrong way. yay for my probably horribly inefficient vector-fold
06:26 (join) si14
06:59 (join) sstrickl
07:06 (join) RacketCommitBot
07:06 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://bit.ly/qiJLQ5
07:06 RacketCommitBot: [racket/master] add more info to try to debug planet failure - Robby Findler
07:06 (part) RacketCommitBot
07:08 (quit) DGASAU: Ping timeout: 260 seconds
07:12 (join) DGASAU
07:20 (join) RacketCommitBot
07:20 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://bit.ly/qPN7uI
07:20 RacketCommitBot: [racket/master] move some helper files into the private/ directory for drracket tests - Robby Findler
07:20 (part) RacketCommitBot
07:50 (join) dnolen
08:23 (join) RacketCommitBot
08:23 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://bit.ly/qEIudW
08:23 RacketCommitBot: [racket/master] adjust sandbox to give S-expression input "original" status - Matthew Flatt
08:23 (part) RacketCommitBot
08:25 (quit) MayDaniel: Read error: Connection reset by peer
09:05 (join) chturne
09:08 (join) jeapostrophe
09:21 (part) lewis1711
09:33 (join) lucian
09:36 (join) ChibaPet
09:38 (quit) noelw: Quit: noelw
09:43 (quit) dnolen: Quit: dnolen
10:01 (join) RacketCommitBot
10:01 RacketCommitBot: [racket] plt pushed 3 new commits to master: http://bit.ly/nHoZEH
10:01 RacketCommitBot: [racket/master] Fix a scribble test for plural "Examples" - Eli Barzilay
10:01 RacketCommitBot: [racket/master] Typo in error message. - Eli Barzilay
10:01 RacketCommitBot: [racket/master] When a file is missing the ftp client throws an error, catch it. - Eli Barzilay
10:01 (part) RacketCommitBot
10:06 (join) anRch
10:06 (quit) realitygrill: Quit: realitygrill
10:35 (quit) hkBst: Read error: Connection reset by peer
10:40 (quit) ChibaPet: Quit: Leaving.
10:41 (join) ChibaPet
10:53 (quit) chturne: Quit: Leaving
10:54 (quit) jeapostrophe: Quit: jeapostrophe
11:04 (join) dnolen
11:07 ChibaPet: same
11:14 (quit) stride: Ping timeout: 264 seconds
11:14 (join) stride
11:26 (join) RacketCommitBot
11:26 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://bit.ly/oQlrDS
11:26 RacketCommitBot: [racket/master] Use a temporary directory to eliminate DrDr races in running benchmarks. - Sam Tobin-Hochstadt
11:26 (part) RacketCommitBot
11:29 (quit) anRch: Quit: anRch
11:39 (join) RacketCommitBot
11:39 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://bit.ly/poOjHc
11:39 RacketCommitBot: [racket/master] Move the "other editors" section later in the guide. - Vincent St-Amour
11:39 (part) RacketCommitBot
11:47 danking: So this is a macro stumbling block I've ran into a few times. I have a pattern variable which was matched as (foo ...) and I'd like to use the elipses to produce a series of expressions where each foo has been modified a bit.
11:48 danking: Specifically, I want to write a DEFINE-USTRUCT form which will expand into a normal (struct ...) expression and a series of (define (structname-update-fieldname s fld) ...)
11:49 danking: The bodies of the update procedures is fairly trivial, it's just a call to `struct-copy'
11:49 (join) jeapostrophe
11:49 danking: My attempt is here: https://gist.github.com/1151824 .  The error is "missing ellipses with pattern variable in template in: fields"
11:49 danking: I'm not sure how this should be done.
11:50 stamourv: danking: Line 12, you're using `fields' without ellipses. The ellipses on line 14 don't count, since they're in the outer #`.
11:51 stamourv: You could map over the list (either `syntax-map' from `unstable/syntax', or `map')
11:51 stamourv: (over 'syntax->list').
11:52 danking: I was hoping to not use map :/
11:53 stamourv: Why? `map' is nice.
11:59 danking: The pattern matching stuff seems more elegant. The output of my macro looks like how I'd write it if I was explaining it to someone.
12:00 stamourv: Pure templating is nice, but sometimes you do need the expressiveness of an actual programming language.
12:01 stamourv: Also, you may want to look at `syntax-parse'. I never use `syntax-case' anymore, `syntax-case' is much nicer for most things.
12:02 stamourv: For simple cases, you use it pretty much exactly like `syntax-case', but if you need more flexibility, you don't have to rewrite.
12:04 danking: stamourv: Ok, I've got a second attempt, but I don't have any idea what this error means: https://gist.github.com/1151824
12:05 (join) mattmight
12:06 stamourv: danking: You can have multiple fields, right?
12:07 danking: Yeah.
12:07 danking: stamourv: I think it was a repl thing, it seems to be working after I restarted the repl.
12:07 stamourv: Ok.
12:08 danking: Ah excelent, it works now.
12:08 danking: thanks stamourv
12:21 (join) RacketCommitBot
12:21 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://bit.ly/r6Vpr9
12:21 RacketCommitBot: [racket/master] Fix DrRacket tests. - Vincent St-Amour
12:21 (part) RacketCommitBot
12:29 (quit) tfb: Quit: sleeping
12:29 samth: jeapostrophe: is it possible to get the drdr data somewhere?
12:30 jeapostrophe: what data_?
12:30 (join) shofetim
12:30 jeapostrophe: the timing data is linked from each file
12:30 samth: yeah, i'm thinking of the failed/stderr/etc data
12:31 samth: also, we should try to use some JS graphing library to do the timing graphs
12:31 samth: take the load off of the server
12:31 jeapostrophe: do you want it per-file or a big thing of a whole version?
12:31 samth: the big thing
12:31 samth: i'd like to figure out which files consistently fail
12:31 samth: so that we can disable them
12:34 jeapostrophe: well, in that case, the best thing to do is give you access to the machine so you can write a script using the files directly
12:34 samth: ok
12:34 jeapostrophe: all the data right now is 12G
12:34 samth: how big a single push worth of data?
12:34 jeapostrophe: about 80M
12:35 jeapostrophe: but it gets compressed once it isn't in the top 150 pushes
12:35 jeapostrophe: and then is just 30M
12:37 jeapostrophe: so if you want to use the machine, email me a pub key
12:43 (join) MayDaniel
12:55 (quit) martinhex: Read error: Connection reset by peer
12:58 (quit) Demosthenes: Quit: Lost terminal
13:05 (quit) MayDaniel: Read error: Connection reset by peer
13:07 (join) RacketCommitBot
13:07 RacketCommitBot: [racket] plt pushed 2 new commits to master: http://bit.ly/nVwlMj
13:07 RacketCommitBot: [racket/master] Fixes indentation - Casey Klein
13:07 RacketCommitBot: [racket/master] Adds a syncheck:add-arrow example to the docs - Casey Klein
13:07 (part) RacketCommitBot
13:22 (join) jonrafkind
13:33 (join) weirdo
13:34 weirdo: hello, can i make mpairs display exactly as pairs in the pretty printer?
13:34 weirdo: please, please, please, i need this for my r5rs endeavors
13:37 weirdo: or make it possible to disable 'mcons' printing no matter what
13:41 weirdo: esp. with print-as-expression
13:54 (join) realitygrill
14:03 (quit) lucian: Read error: Operation timed out
14:09 (join) martinhex
14:24 samth: weirdo: see the `print-mpair-curly-braces' parameter
14:24 samth: rudybot: doc print-mpair-curly-braces
14:24 rudybot: samth: your sandbox is ready
14:24 rudybot: samth: http://docs.racket-lang.org/reference/Writing.html#(def._((quote._~23~25kernel)._print-mpair-curly-braces))
14:24 weirdo: samth, but it prints as (mcons a b)
14:24 weirdo: when print-as-expression is true
14:25 samth: then turn off print-as-expression
14:26 weirdo: rudybot, doc print-as-expression
14:26 rudybot: weirdo: your r5rs sandbox is ready
14:26 rudybot: weirdo: no docs for a current binding, but provided by: racket/base, racket
14:27 samth: rudybot: doc print-as-expression
14:27 rudybot: samth: your sandbox is ready
14:27 rudybot: samth: http://docs.racket-lang.org/reference/Writing.html#(def._((quote._~23~25kernel)._print-as-expression))
14:27 weirdo: thanks
14:27 samth: however, using r5rs shouldn't do any of this
14:27 samth: rudybot: init r5rs
14:27 rudybot: samth: your r5rs sandbox is ready
14:28 samth: rudybot: eval (display (cons 1 2))
14:28 rudybot: samth: ; stdout: "(1 . 2)"
14:28 samth: weirdo: look, no curly braces or mcons
14:28 weirdo: samth, i'm using geiser-mode in emacs, which is written in racket
14:29 weirdo: but my program is in r5rs
14:29 samth: (a) geiser is written in elisp, but it's *for* racket
14:29 samth: (b) what program are you running, and what output are you seeing?
14:29 weirdo: geiser has a 'server' running on Racket
14:30 weirdo: samth, expression '(1 2 3), i see something along the lines of (mcons 1 (mcons 2 (mcons 3 ()))
14:31 weirdo: after doing ,enter mylib in geiser
14:31 weirdo: doesn't print-as-expression give prettier output?
14:32 samth: then geiser is using racket's `print' function, not the r5rs `print' function
14:33 weirdo: oh. so current-print needs to be bound to r5rs-print, right?
14:34 weirdo: thank you, it's very enlightening :-)
14:35 samth: that might be the right thing to do
14:35 samth: alternately, you could change the settings of `print-as-expression'
14:35 samth: and `print-mcons-curly-braces'
14:37 (join) prototrout
14:39 (join) lucian
14:47 weirdo: thank you. it all works now
14:47 (quit) masm: Quit: Leaving.
14:53 (quit) si14: Ping timeout: 250 seconds
15:00 (join) lucian_
15:02 (quit) lucian: Ping timeout: 264 seconds
15:16 (part) prototrout
15:23 (join) MayDaniel
15:26 (join) soegaard
15:32 (quit) soegaard: Ping timeout: 252 seconds
15:43 (quit) MayDaniel: Read error: Connection reset by peer
15:47 (join) hussaibi
16:04 ohwow_: what's algebraic stepper?
16:16 (join) RacketCommitBot
16:16 RacketCommitBot: [racket] plt pushed 2 new commits to master: http://bit.ly/q9aqM6
16:16 RacketCommitBot: [racket/master] [honu] handle unary operators. add some common functions and binary operators - Jon Rafkind
16:16 RacketCommitBot: [racket/master] [honu] dont remove syntaxes that end expressions from the stream so macros can consume them. add a simple cond macro - Jon Rafkind
16:16 (part) RacketCommitBot
16:19 (join) dsrguru
16:23 samth: ohwow_: a debugger that represents the algebraic transformations of the language semantics directly as reduction steps
16:23 dsrguru: I want to write a macro (for/text ([id lst] ...) body) that duplicates body (length lst) many times, replacing occurrences of id in body with elements of lst
16:23 dsrguru: without executing body
16:24 samth: dsrguru: I'm confused -- can you give an example and what it should expand to?
16:24 dsrguru: sure
16:25 dsrguru: (map somefunc (for/text ([str '("one" "two" "three")]) (format "number ~a" str)))
16:25 dsrguru: should become
16:25 dsrguru: (map somefunc (format "number ~a" "one") (format "number ~a" "two") (format "number ~a" "three"))
16:26 samth: dsrguru: that's not possible -- you can't expand into multiple forms
16:27 samth: also, how will you know how long `lst' is?
16:27 stamourv: dsrguru: Should the 3 `format' be in a list?
16:27 samth: (at macro-expansion time, that is)
16:27 dsrguru: I was planning to recursively call the macro
16:27 dsrguru: until lst was null
16:27 samth: but `lst' is an expression, right?
16:27 dsrguru: stamourv: no, I was hoping there was a way to produce multiple forms
16:27 dsrguru: samth: yes
16:28 samth: then how could it possibly know at expansion time how long `lst' is?
16:28 samth: what about: (for/text ([str (read)]) ...)
16:28 dsrguru: I guess that wouldn't be possible :)
16:28 dsrguru: I'm new to the lisp way of doing things
16:28 dsrguru: and macros seem magically powerful
16:29 samth: what larger goal are you trying to accomplish?
16:29 dsrguru: I have a map call
16:29 dsrguru: where the procedure takes many arguments
16:29 dsrguru: and each list that I pass to map
16:29 dsrguru: is a very similar function call
16:29 dsrguru: but with one parameter different
16:29 samth: how about this:
16:29 dsrguru: as in the above map example
16:30 samth: rudybot: eval (apply map string-append (for/list ([str '("one" "two" "three")]) (list (format "number ~a" str)))))
16:30 rudybot: samth: error: eval:1:36: read: illegal use of open square bracket
16:30 dsrguru: (I considered applying map to the appended form)
16:30 samth: rudybot: init racket
16:30 rudybot: samth: your sandbox is ready
16:30 samth: rudybot: eval (apply map string-append (for/list ([str '("one" "two" "three")]) (list (format "number ~a" str)))))
16:30 rudybot: samth: error: eval:1:99: read: unexpected `)'
16:30 samth: rudybot: eval (apply map string-append (for/list ([str '("one" "two" "three")]) (list (format "number ~a" str))))
16:30 rudybot: samth: ; Value: ("number onenumber twonumber three")
16:31 samth: dsrguru: does that work?
16:31 dsrguru: no
16:31 dsrguru: in this example
16:31 dsrguru: I'd need
16:31 dsrguru: :
16:31 dsrguru: (map somefunc "number one" "number two" "number three")
16:31 dsrguru: so instead of string-append
16:31 samth: that is what you got
16:31 dsrguru: regular append
16:32 dsrguru: wait
16:32 dsrguru: oh
16:32 samth: map takes lists as arguments -- not strings
16:32 dsrguru: you mean string-append was just the func
16:32 dsrguru: you chose
16:32 samth: yes
16:32 dsrguru: ok
16:32 dsrguru: so the problem in my example
16:32 dsrguru: is that
16:32 dsrguru: not all of the lists
16:32 samth: rudybot: eval (apply map list (for/list ([str '("one" "two" "three")]) (list (format "number ~a" str))))
16:32 rudybot: samth: ; Value: (("number one" "number two" "number three"))
16:32 dsrguru: are to be generated
16:32 dsrguru: by the for comprehension
16:32 dsrguru: so
16:32 dsrguru: I forgot to mention that
16:32 samth: rudybot: eval (apply map list (list 'the-first-element) (for/list ([str '("one" "two" "three")]) (list (format "number ~a" str))))
16:33 rudybot: samth: ; Value: ((the-first-element "number one" "number two" "number three"))
16:33 dsrguru: oh
16:34 dsrguru: thanks
16:35 dsrguru: samth: yeah that's exactly what I want
16:36 dsrguru: I didn't realize apply worked that way; I thought I'd have to append the lists I already had to the result of for/list and pass that whole newly computed list to apply
16:37 dsrguru: oh I misread that
16:37 dsrguru: you're essentially doing that by mapping the output of for/list back into a new list
16:38 dsrguru: I guess mapping the list is more readable
16:44 dsrguru: samth: so the main thing for me to take away from this is that macros, like functions, produce a single expression, not arbitrary text
16:44 (quit) dnolen: Ping timeout: 252 seconds
16:48 jonrafkind: yes, you replace one expression with another
16:48 jonrafkind: but of course you can have subexpressions
16:49 (join) soegaard
16:54 samth: dsrguru: yes, that's correct -- you can use `begin' to do multiple things, but you can't produce more than one form
16:56 dsrguru: all right thanks
16:57 (quit) dsrguru: Quit: leaving
16:57 (join) dnolen
17:12 samth: rudybot: init typed/racket
17:12 rudybot: samth: your typed/racket sandbox is ready
17:12 samth: rudybot: eval (apply (λ: ([x : String] [y : String]) "") (list "foo" "bar"))
17:12 rudybot: samth: error: evaluator: terminated (out-of-memory)
17:13 samth: offby1: i think your memory limit is too low
17:14 ohwow_: hm
17:14 ohwow_: this looks like a simple program, why would it use much memory?
17:16 samth: the typechecker uses a bunch of memory
17:16 samth: rudybot: help
17:16 rudybot: samth: help [<what>], version, quote, source, seen <nick>, uptime, t8 <from> <to> <text> ..., init [<lang>], eval <expr> ..., give <to> <expr> ..., apropos <str> ..., desc <id>, doc <id>, later "tell" <nick> <something> ...
17:28 (part) ChibaPet
17:35 (quit) mceier: Quit: leaving
17:35 (join) RacketCommitBot
17:35 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://bit.ly/rcXLt7
17:35 RacketCommitBot: [racket/master] Fix DrRacket test. - Vincent St-Amour
17:35 (part) RacketCommitBot
17:47 (quit) hussaibi: Ping timeout: 260 seconds
17:51 offby1: samth: tell me about it; I can never remember anything :)
17:52 samth: offby1: have you considered a career doing standup?
17:53 offby1: isn't that what I _am_ doing?
17:53 offby1: it's 20 megabytes
17:55 samth: offby1: i expect that's too small for TR at all
17:55 samth: rudybot: eval 1
17:55 rudybot: samth: your typed/racket sandbox is ready
17:55 rudybot: samth: ; Value: 1
17:56 samth: rudybot: eval (+ 1 2)
17:56 rudybot: samth: ; Value: 3
17:56 samth: rudybot: eval (lambda (x) x)
17:56 rudybot: samth: ; Value: #<procedure>
17:56 (join) Demosthenes
17:56 samth: offby1: i thought it included the results of stdout?
17:56 samth: eval (apply (λ: ([x : String] [y : String]) "") (list "foo" "bar"))
17:56 samth: rudybot:  eval (apply (λ: ([x : String] [y : String]) "") (list "foo" "bar"))
17:56 rudybot: samth: error: eval:1:0: Type Checker: Bad arguments to function in apply: <NEWLINE> Domain: String String <NEWLINE> Arguments: (List String String) * <NEWLINE>  in: (#%app apply (lambda (x y) (quote "")) (#%app list (quote "foo") (quote "bar")))
17:57 samth: ah, there it goes
17:59 offby1: the box in which it runs (an ec2 "micro" instance)  has only about 150 Mb free, and no swap
17:59 (join) anRch
17:59 samth: that is pretty micro
17:59 samth: any idea why it failed the first time, and ran the second time?
18:02 (quit) jeapostrophe: Quit: jeapostrophe
18:06 (quit) sstrickl: Quit: sstrickl
18:06 (quit) dnolen: Quit: Page closed
18:08 offby1: noe wahtever
18:09 (join) rekahsoft
18:09 rekahsoft: hi all...was wondering how i would go about using provide/contract with a struct that uses auto values..
18:13 rekahsoft: for example with a struct define like so: (struct posn ([x #:auto] [y #:auto]) #:auto 0 #:transparent) how would the provide/contract look?
18:15 rekahsoft: nvm i found the issue..
18:16 rekahsoft: lol taken back..don;t think anyone is there tho :(
18:17 samth: rekahsoft: we're here, but you can't expect an instant answer
18:17 samth: also, i don't think you found the answer -- it doesn't work for me
18:20 (join) RacketCommitBot
18:20 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://bit.ly/quEqcE
18:20 RacketCommitBot: [racket/master] Handle apply for functions with no rest arguments. - Sam Tobin-Hochstadt
18:20 (part) RacketCommitBot
18:24 rekahsoft: samth: i just stated that ^
18:24 rekahsoft: and what doesn't work for you?
18:24 (quit) soegaard: Quit: Page closed
18:26 samth: what did you write in `provide/contract', rekahsoft?
18:26 samth: this always fails for me:  (provide/contract (struct posn ([x integer?] [y integer?])))\
18:27 samth: because the `posn' constructor takes 0 arguemtns, so the contract errors
18:27 (nick) samth -> samth_away
18:27 samth_away: have to run
18:29 rekahsoft: samth_away: alright :( and i have the same thing as my provide/contract..i just don't know how to get around this auto-values issue
18:53 (quit) anRch: Quit: anRch
19:04 (join) masm
19:05 (join) jeapostrophe
19:29 (nick) Lajla -> Geisteskrankh
19:37 (quit) masm: Quit: Leaving.
19:50 (quit) DGASAU: Remote host closed the connection
19:53 (join) DGASAU
19:59 (join) bmp
20:00 (quit) realitygrill: Quit: realitygrill
20:25 (join) lewis1711
20:28 (quit) bmp: Remote host closed the connection
20:28 (join) bmp
21:23 (quit) jonrafkind: Ping timeout: 260 seconds
21:45 (quit) jeapostrophe: Quit: jeapostrophe
21:45 (join) hussaibi
22:44 (join) WarWeasle
22:45 WarWeasle: Hello. I've been trying to compile a simple racket/gui app (based on the examples) but when I complile and run them on windows, I get a console window as well. What am I doing wrong?
22:51 eli: samth_away: the memory accounting that custodians provide can be unpredictable.
22:51 eli: offby1: I'd second Sam's career suggestion...
22:52 (join) jeapostrophe
22:53 WarWeasle: Am I connected?
22:54 eli: WarWeasle: Yes.
22:54 WarWeasle: eli: Thank you. I spent over an hour once before nickserv oked me.
22:54 (quit) jeapostrophe: Client Quit
22:56 eli: WarWeasle: registering with it should avoid that, but I've never heard of problems with random nicks.  Maybe they'd gotten more strict though.
22:57 WarWeasle: eli: It was a chanel that required registered nicks, and I had just reinstalled emacs so I didn't have auto-register working yet.
22:57 lewis1711: is there a way to generate a large list like iota in srfi1, or [m..n] in haskell?
22:58 WarWeasle: Any Windows users around?
23:00 eli: WarWeasle: Ah; no, we don't require that -- we even have this thing: http://racket-lang.org/irc-chat.html
23:00 ohwow_: (Sorry for offtopic, but this might be interesting http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1773169 for some people here )
23:00 eli: rudybot: (length (for/list ([x (in-range 10000)]) x))
23:00 rudybot: eli: your sandbox is ready
23:00 rudybot: eli: ; Value: 10000
23:00 eli: lewis1711: ^
23:01 eli: rudybot: (length (sequence->list (in-range 10000)) x)
23:01 rudybot: eli: error: reference to an identifier before its definition: x in module: 'program
23:01 eli: rudybot: (require racket/list)
23:01 rudybot: eli: Done.
23:01 eli: rudybot: (length (sequence->list (in-range 10000)) x)
23:01 rudybot: eli: error: reference to an identifier before its definition: x in module: 'program
23:01 eli: rudybot: (length (sequence->list (in-range 10000)))
23:01 rudybot: eli: ; Value: 10000
23:01 eli: lewis1711: That too.
23:02 lewis1711: eli: aha, I came across in-range but didn't figure it out. thanks
23:04 WarWeasle: ohwow_: That IS interesting. I'll have to read it later.
23:04 (join) RacketCommitBot
23:04 RacketCommitBot: [racket] plt pushed 2 new commits to master: http://bit.ly/peyoBH
23:04 RacketCommitBot: [racket/master] fix plurality of `examples' label - Matthew Flatt
23:04 RacketCommitBot: [racket/master] Revert "Fix a scribble test for plural "Examples"" - Matthew Flatt
23:04 (part) RacketCommitBot
23:09 WarWeasle: Does anyone use Racket for Windows GUI development? I'd like to try it, but I'm getting a console window along with my GUI when I compile the examples to exe. Any ideas what I'm doing wrong?
23:09 (join) asurai
23:10 ohwow_: WarWeasle: have you added --gui option?
23:10 ohwow_: raco exe --gui test.rkt
23:11 eli: WarWeasle: Yes, you need the gui option.
23:11 WarWeasle: ohwow_: No, I was using DrRacket and selecting all the permutations.
23:11 WarWeasle: Of compile.
23:11 ohwow_: eh
23:12 ohwow_: i am not really familiar with DrRacket, let me open it up first
23:12 eli: WarWeasle: Note that there's racket.exe and gracket.exe -- the latter is not really needed, since racket.exe can do everything, but an executable on windows need to identify itself as a gui or a console app, which is the only difference between the two.
23:12 eli: There should be an option in the drracket menu thing too.
23:14 WarWeasle: eli: But I can simply use raco if there isn't?
23:14 (quit) bmp: Ping timeout: 258 seconds
23:16 WarWeasle: Also, what is the proper way to use multiple files to create a project.
23:16 lewis1711: https://gist.github.com/1153214 there's probably stuff wrong with this, but using comprehensions for fold seems to be consistently faster than the built in list function. and generic to boot!
23:17 (quit) asurai: Quit: Bye!
23:29 (quit) blomqvist: Ping timeout: 260 seconds
23:30 (quit) hyko: Read error: Connection reset by peer
23:30 (join) hyko
23:30 eli: lewis1711: no, I get a slightly faster time for the first time.
23:30 eli: lewis1711: no, I get a slightly faster time for the first `time'.
23:31 eli: WarWeasle: Yes.  (But there should be.)
23:31 (quit) WarWeasle: Remote host closed the connection
23:33 lewis1711: eli: I get consistently slightly faster for the last time, unless I add another 0 to the list generator thing. I think I'll just keep it for vector folding (since sequence-fold gives different results)
23:33 eli: lewis1711: It should be pretty similar, usually.
23:34 lewis1711: what should be?
23:34 eli: The thing is that the list form is kind of like inlining the body of `foldl', since it expands to roughly the same code.
23:34 eli: (The first and the last should be roughly the same.)
23:34 lewis1711: ah
23:35 eli: So the differences are possibly all in the noise, and can fluctuate on different list sizes, whether the GC happens to kick in in one of them or not, etc.
23:35 lewis1711: out of curiosity, do you know why sequence-fold gives different results than foldl? I think someone expained it before to me here (possibly even you) but I forgot the rationale
23:36 (join) blomqvist
23:38 (join) jonrafkind
23:40 (join) realitygrill
23:45 eli: lewis1711: it's similar to the difference that you'll see between
23:46 eli: (for ([x (in-range 1000)]) blah)
23:46 eli: and
23:46 eli: (let ([r (in-range 1000)]) (for ([x r]) blah))
23:46 eli: In the first case, the `for' recognizes `in-range' and generates code, but in the second case it generates a value that the for loop needs to work with.
23:48 (quit) hussaibi: Ping timeout: 252 seconds
23:49 lewis1711: hmm, that's beyond me :(
23:50 eli: lewis1711: Think about it this way.
23:50 eli: When `for' sees `in-range', as in (for ([x (in-range 1000)]) blah)
23:50 eli: it knows the kind of code that it needs to generate.
23:50 eli: You can actually use the macro stepper and see how it constructs a named-let loop that is roughly similar to what you'd write.
23:51 eli: But if you first do (define r (in-range 1000)), then r is bound to a value.
23:51 eli: Now, when you do (for ([x r]) blah), the `for' loop has no idea what the range is.
23:51 eli: All it knows, is that it's a "sequence".
23:51 eli: So it uses generic sequence operations to iterate over it.
23:52 eli: And that's very different from the near-hand-written loop in the previous case.
23:52 lewis1711: hmmm. that sort of makes sense
23:52 eli: You can try both with the macro stepper, and see that the two `for' forms expand to very different code.
23:53 lewis1711: I will. didn't even know about the macro-stepper, looks handy
23:56 lewis1711: it's not being picked up by the macro stepper