00:07 (join) jonrafkind
00:28 (quit) dnolen: Quit: dnolen
01:07 (quit) veer: Quit: Leaving
01:14 (quit) realitygrill: Quit: realitygrill
01:23 (quit) jeapostrophe: Quit: jeapostrophe
01:29 jao: danking, eli, i'm afraid xrepl doesn't work in geiser.  i'm using my own unquoting mechanism and a customized repl there.
02:18 (quit) jonrafkind: Ping timeout: 260 seconds
02:24 (join) dmac1
02:28 mithos28: Is there a way to get set! to support l-values?
02:33 (join) hkBst
02:38 (quit) wtetzner: Ping timeout: 248 seconds
02:43 (join) wtetzner
02:45 (quit) dmac1: Ping timeout: 258 seconds
03:24 (join) dmac1
03:24 (nick) elliottcable -> elli
03:28 (nick) elli -> elliottcable
03:32 (join) ahinki
03:33 (join) tim-brown
03:33 (join) Blkt
03:36 (quit) mithos28: Quit: mithos28
03:37 (nick) elliottcable -> llttcbl
03:38 Blkt: good morning everyone
03:45 (nick) llttcbl -> elliottcable
03:49 tim-brown: mornin Blkt, mornin all
03:49 Blkt: :)
03:50 tim-brown: llttcbl: i see you've joined #vim. do you use embedded :mz in vim?
04:00 (quit) slilo: Ping timeout: 258 seconds
04:19 (join) masm
04:21 (join) bluezenix
04:25 (join) sid0
04:26 sid0: Hi, I'm new to Racket and have what seems to be a newbie question on macros
04:28 (quit) karswell: Read error: Operation timed out
04:29 (join) karswell
04:29 sid0: http://paste.lisp.org/display/125262
04:30 sid0: What seems to be happening is that the fname in the with-syntax pattern doesn't get replaced with bar
04:33 sid0: I'm probably missing something pretty basic here :)
04:48 sid0: Sorry, I guess it's not really clear enough from my example. Here's a more involved example: http://paste.lisp.org/display/125263
04:49 sid0: I'm trying to write an unhygienic macro following from http://calculist.blogspot.com/2008/03/intentional-capture.html
04:50 sid0: which plops my function in for the given one in the syntax's lexical context
05:36 hkBst: sid0: you're syntax-binding fname1, but you're calling fname2..., plus you're trying to bind a symbol, instead of the argument to your macro, see my annotation to your second paste
05:36 (quit) xharkonnen_: Quit: Leaving
05:39 sid0: hkBst: thanks, but that's still printing out bar
05:40 hkBst: sid0: yes, looking closer at your source, I must admit, I'm unclear what you're trying to accomplish
05:40 sid0: hkBst: yeah I realize my example sucks
05:41 sid0: coming up with a better one now
05:46 sid0: hkBst: ooh
05:47 sid0: hkBst: http://paste.lisp.org/display/125268 is a better example of what I'm trying to do
05:47 sid0: hkBst: but I see you mentioned the solution already, since changing the second argument of (datum->syntax #'op 'replace-id) to #'replace-id causes it to work
05:48 sid0: hkBst: but I'm not sure why, since isn't the second argument of datum->syntax supposed to be a datum and not a syntax object?
05:53 hkBst: sid0: yes, in this new macro I can see it makes more sense this way :)
05:53 eli: jao: Any inherent reason that it won't work?  It's a pity that both tools are so close but they won't work together.
05:53 eli: sid0: What are you trying to do, exactly?  (Description, not these examples...)
05:54 sid0: eli: given a lambda with a call to another function inside the lambda I want to replace the callee with my own function
05:55 eli: sid0: And the name that you're trying to replace is always the same?
05:55 sid0: No, I'd like to pass that in as an argument to the macro
05:56 sid0: I had managed to get it to work for the constant case
05:56 sid0: but I was having trouble generalizing that
05:56 eli: Um, if you pass it in as an argument, then there is no need for breaking hygiene in any way...
05:56 sid0: as an argument to the *macro*, not the lambda :)
05:57 sid0: the lambda's already specified and the function callinside's bound beforehand
05:57 eli: Just rewrite (foo old new (lambda (x ...) body ...)) -> (lambda (x ...) (let ([old new]) body ...))
05:57 jao: eli, i use my own evaluator to introduce meta-commands that are used by geiser to interact with racket from code buffers, and to keep track of modules loaded and such (i have a slightly modified version of enter!)
05:58 sid0: eli: Ah, I remember trying that but not getting it to work
05:58 sid0: eli: I should try it again I guess
05:58 eli: sid0: Yeah, that would be more useful.
05:59 eli: jao: Then why not make geiser just use xrepl directly then?
05:59 jao: eli, i could take xrepl's as a starting point to my extensions,
05:59 eli: jao: I'd be happy to help if there's anything that is needed...
05:59 jao: eli, yes, exactly.  just a matter of finding a time slot.
05:59 eli: BTW, what's different in your `enter!' variant?
06:00 jao: eli, thanks.  i'll take a look.  i'd be actually happy to take advantage of your work
06:00 eli: jao: And the `enter!' question?
06:00 jao: eli, iirc, i'm keeping track of module loading in my enter!, to avoid loading modules when doing the autodoc thing
06:01 eli: Why is that needed?
06:01 eli: If a module is loaded you won't pay the price for reloading it, so that shouldn't be a problem.
06:01 jao: eli, i don't want modules to be loaded unless the user explicitly asks for them
06:02 jao: or, at least, i didn't want back in the day :)
06:02 eli: I'm not following how that's related to entering a module...
06:02 eli: Perhaps you're trying to know which modules are currently instantiated?
06:02 sid0: eli: yeah, it works :) now to figure out what I was doing wrong then
06:03 eli: sid0: If you're new to Racket and coming from CL, it's best to just avoid the temptation to try unhygienic macros.
06:03 jao: eli, i'll have to look at the code.  i forgot the details.
06:03 eli: sid0: If you do really want one, then have a look at syntax parameters.
06:03 sid0: eli: will this work for an arbitrary expression and not just a lambda?
06:03 eli: sid0: It should, as long as you know where to place the `let' (which you need with the macro anyway).
06:04 eli: jao: If you need something along these lines, then there's code that can work better which I put in xrepl, which is why I think that it's best if you just use it.
06:05 jao: eli, fair enough.  i'll see what i can do.
06:05 eli: jao: BTW, I plan on adding a few more guile-isms, IIRC, one thing that looked useful was the ",in" thing.
06:06 jao: eli, as a somehow related aside, any plans to extend help to just provide the URLs instead of directly launching a browser (as we discussed some time ago in the ML)?
06:06 sid0: eli: Yeah, seems to work fine. Thanks! Yeah, coming from CL hygienic macros are a little mind-bending
06:06 sid0: I'm really used to being able to simply replace names
06:07 eli: jao: Yeah, I still have that on my todo list...  My plan was to keep `help' the same, and add a `help*' that is similar but evaluates to the url.
06:07 jao: eli, re guile-isms: that sounds great, and yet another reason to use xrepl.  i'm guessing it provides a way to define my own meta-commands? (that's how it works for guile, and actually ,in was a request of mine's)
06:07 eli: (I have some of that work done and waiting on a git stash...)
06:08 jao: ah, excellent
06:08 eli: jao: There's no "published" api for adding commands -- but that's just a technicality.
06:08 eli: Since it's new, I didn't want to commit on a particular api, so I kept it out.
06:08 jao: eli, as long as there's a way, published or not, i'll be fine
06:08 eli: But the file certainly has a `defcommand' macro which I could providein some way.
06:09 eli: Yeah, there is a way...
06:09 jao: okay, expect some nagging in the not too far future then ;)
06:10 eli: jao: Sure.  FWIW, the text I wrote somewhere is that you could do:
06:10 eli: (eval '(defcommand eli "stuff" "eli says" ["Make eli say stuff"] (printf "Eli says: ~a\n" (getarg 'line))) (module->namespace 'xrepl/xrepl))
06:10 eli: and dump that in your ~/.racketrc
06:11 eli: Ugly, but intentionally so -- just in case someone really wants to extend it and can't wait for a proper api...
06:11 eli: sid0: It will help a lot to avoid that -- when you see a case where you have to do it, then ask.  Chances are that you don't need it.
06:12 eli: sid0: If you want, we've recently wrote a paper on syntax parameters, which has a bunch of mid-level macro discussions about hygiene, and it might be helpful to just read through it.
06:30 sid0: eli: A link to the paper would be good in case I do need them at some point
06:39 sid0 finds http://blog.racket-lang.org/2008/02/dirty-looking-hygiene.html -- pretty cool
07:10 eli: sid0: http://barzilay.org/misc/stxparam.pdf
07:17 (quit) sid0: Ping timeout: 252 seconds
07:23 (join) jeapostrophe
07:23 (quit) jeapostrophe: Client Quit
07:48 (join) mceier
07:58 (join) dnolen
08:00 (quit) mceier: Ping timeout: 260 seconds
08:03 (join) mceier
08:10 (quit) mceier: Read error: Operation timed out
08:25 (join) mceier
08:31 (quit) dmac1: Ping timeout: 244 seconds
09:37 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/QEylyw
09:37 RacketCommitBot: [racket/master] document unexpected mouse event reporting; Closes PR 12278 - Matthias Felleisen
10:10 (quit) dnolen: Quit: dnolen
10:35 (nick) samth_away -> samth
10:44 samth: tewk: that (sending the messages back) won't work -- i'd have to change the code running in every place the program creates to do that
10:47 (quit) hkBst: Remote host closed the connection
10:47 (quit) ahinki: Quit: ChatZilla 0.9.87 [Firefox 8.0/20111006182035]
10:51 (join) anRch
10:55 (join) shofetim
11:38 (quit) tim-brown: Quit: Leaving
11:44 (quit) anRch: Quit: anRch
11:51 (join) anRch
11:53 (join) mithos28
12:11 (join) wishbone4
12:14 (nick) petey` -> petey
12:16 (quit) anRch: Quit: anRch
12:17 (join) bluezenix1
12:17 (quit) bluezenix1: Client Quit
12:18 (quit) bluezenix: Ping timeout: 252 seconds
12:20 (quit) mithos28: Quit: mithos28
12:20 (quit) Blkt: Quit: going home
12:28 (join) xharkonnen
12:52 (join) dnolen
13:44 (join) anRch
13:44 (join) jonrafkind
13:50 (quit) anRch: Ping timeout: 252 seconds
13:51 (join) anRch
13:52 (quit) masm: Ping timeout: 255 seconds
13:57 (join) masm
14:02 (join) dmac1
14:15 (join) MayDaniel
14:15 (join) bfulgham
14:16 (quit) MayDaniel: Read error: Connection reset by peer
14:31 (join) JoelMcCracken
14:55 (join) mithos28
14:57 (quit) anRch: Quit: anRch
15:12 (join) DanBurton
15:13 (quit) dnolen: Ping timeout: 252 seconds
15:32 (join) Thelas
15:33 (nick) Thelas -> Staloras
15:41 Staloras: What higher-order function should I be using to write something like this without explicit recursion?
15:41 Staloras: (define permute-real (lambda (alat)                        (let ((a (car alat)) (lat (cdr alat)))                   (cond                     ((null? (cdr lat)) (fill a (car lat)))                     (else (append (fill a (car lat)) (permute-real a (cdr lat))))))))
15:41 Staloras: wait, that looks horrible.  Here.  http://codepad.org/lndJSeEd
15:43 stamourv: Staloras: Is that homework?
15:47 (quit) bfulgham: Quit: bfulgham
15:47 shofetim: Staloras: http://codepad.org/G98HRjN4 just prettier, I don't know the answer to your question
15:47 shofetim: 
15:48 (quit) jao: Remote host closed the connection
15:49 JoelMcCracken: Staloras: it would be easier if you could describe what the function is supposed to do
15:51 Staloras: It is an assignment, but if I understand the policy on this correctly... well, the prof actually encouraged me to ask questions on the racket-users group if I need to, and this seems equivalent...  but either way I just saw it... and I'm honestly rather embarrassed I even had to ask that.  I mean, the code doesn't work now, but now it's a different issue I should be able to solve.
15:52 JoelMcCracken: (what is it suposed to do?)
15:52 DanBurton: :) sure never be afraid to ask. We just don't want to spoil the learning experience for you if it is homework
15:52 Staloras: for an unrelated question... why does the stepper show "the-cons" when it's evaluating things on lists instead of "cons" sometimes?
15:52 (join) realitygrill
15:53 Staloras: And it was a helper function for a helper function for something that finds all the permutations of a list.
15:58 Staloras: I'm probably the only person who understands (at least somewhat) the Y combinator and yet fails at simple things like this...
16:00 stamourv: Staloras: Here's a hint: There's repetition between your cond clauses. If you eliminate it, the answer may become more obvious.
16:00 stamourv: Another hint: What does this answer for the empty list?
16:01 Staloras: Like I said, I figured out that function.
16:02 Staloras: I forgot that I could use foldr to build lists.
16:02 stamourv: You did? I ust have missed that part.
16:02 stamourv: s/ust/must/
16:03 Staloras: not your fault, I fail at clarity. I said "but either way I just saw it..."
16:06 stamourv: Oh, I see it now.
16:07 Staloras: Eww.. Okay, I have a correct answer now.
16:07 (join) jao
16:07 Staloras: So now I can ask "is there a way to make this function look less horrible"?
16:09 Staloras: http://codepad.org/MTfBZj1m
16:14 stamourv: Staloras: I recommend line breaks.
16:14 stamourv: Also, these `cons'es can be replaced with quasiquoting and unquoting, which would look nicer.
16:15 Staloras: that's the first time I've ever heard that
16:16 Staloras: people keep asking me why I use `((,a)) over (list (list a))
16:16 stamourv: I use that all the time.
16:21 Staloras: how do I get the (cdar y) part to actually use the elements, though, instead of a list?  [Read: if, say, a is 2, x is 3, and (cdar y) evaluates to 1, (cons a (cons x (cdar y)) is `(2 3 1), but `(,a ,x ,(cdar y)) is, of course, `(2 3 (1))...
16:22 stamourv: Use ,@.
16:22 stamourv: That's unquote-splicing.
16:23 (quit) JoelMcCracken: Remote host closed the connection
16:23 Staloras: Oh, thank you.  That appears to work.
16:28 Staloras: Is unquote-splicing just an equivalent of `(a ,@b) = `(a . ,b)?
16:29 stamourv: rudybot: eval `(1 2 3 ,@'(4 5 6) 7)
16:29 rudybot: stamourv: your sandbox is ready
16:29 rudybot: stamourv: ; Value: (1 2 3 4 5 6 7)
16:30 stamourv: Staloras: No.
16:30 Staloras: Ah.
16:31 stamourv: `(a . ,b) is equivalent to `(a ,@b), but unquote-splicing can do more.
16:31 Staloras: Things are beginning to make sense again.
16:31 Staloras: Though every time I say that something stupid happens..
16:32 Staloras: (by "stupid" I mean "reminding me that I really have no clue how this whole language works")
16:33 Staloras: Like say the part where I saw the Y combinator for the first time.
16:33 Staloras: Or the Y! combinator.
16:33 Staloras: Or the fact that the two aren't actually equivalent.
16:33 stamourv: Take it one bit at a time.
16:33 stamourv: Y is pretty tricky.
16:34 stamourv: And you can certainly write a lot of useful programs without ever needing it.
16:34 Staloras: I understand how those ones work NOW. (I think.)
16:35 Staloras: Just please don't use set! and call-with-current-continuation in the same program.  Whenever those things are combined I can't figure out anything that's happening.
16:41 Staloras: Remind me what happens when you do that, anyways?  It saves basically all the calls AROUND the call to call/cc and when you call whatever you set! equal to the continuation from call/cc it acts as though call/cc evaluated to that and goes from there?
16:41 Staloras: Is that how it works?
16:42 stamourv: Give it a try.
16:42 stamourv: rudybot: Will you give Staloras a hand?
16:42 rudybot: stamourv: LiSP is supposed to be good. I have not read it myself. However, Ikarus might be considered "state of the art" when it comes to a certain class of Scheme compiler. If you like the approach taken by Aziz and Kent Dybvig, their papers are of course, something you should examine. Also see the papers that surround the implementation of Larceny, Stalin, and PLT Scheme.
16:42 stamourv: I'll take that as a yes.
16:45 Staloras: rubybot: (define foo (lambda () (/ 1 0)))) (define bar (lambda () (+ 1 (call-with-current-continuation (lambda (baz) (set! foo baz) 1))))) (bar) (+ 3 (foo (+ 1 1)))
16:45 Staloras: well, typing the name right might help there
16:45 Staloras: rudybot: (define foo (lambda () (/ 1 0)))) (define bar (lambda () (+ 1 (call-with-current-continuation (lambda (baz) (set! foo baz) 1))))) (bar) (+ 3 (foo (+ 1 1)))
16:45 rudybot: Staloras: (begin (define k (call-with-current-continuation values)) (k k))
16:45 stamourv: Staloras: If you give it multiple expressions, you need to tell it to eval explicitly.
16:46 stamourv: Otherwise, it will spit out something that it thinks may be vaguely related.
16:46 stamourv: rudybot: eval (define x 3) x
16:46 rudybot: stamourv: ; Value: 3
16:46 stamourv: Staloras: ^
16:46 Staloras: rubybot: eval (define foo (lambda () (/ 1 0)))) (define bar (lambda () (+ 1 (call-with-current-continuation (lambda (baz) (set! foo baz) 1))))) (bar) (+ 3 (foo (+ 1 1)))
16:46 Staloras: ...
16:46 Staloras: rudybot: eval (define foo (lambda () (/ 1 0)))) (define bar (lambda () (+ 1 (call-with-current-continuation (lambda (baz) (set! foo baz) 1))))) (bar) (+ 3 (foo (+ 1 1)))
16:46 rudybot: Staloras: your sandbox is ready
16:46 rudybot: Staloras: error: eval:1:32: read: unexpected `)'
16:47 Staloras: rudybot: (define foo (lambda () (/ 1 0))) (define bar (lambda () (+ 1 (call-with-current-continuation (lambda (baz) (set! foo baz) 1))))) (bar) (+ 3 (foo (+ 1 1)))
16:47 rudybot: Staloras: yes, please add an alias for call-with-current-continuation (in ze Scheme Prelude/WG1/Small/ hing 1)
16:47 Staloras: rudybot: eval (define foo (lambda () (/ 1 0))) (define bar (lambda () (+ 1 (call-with-current-continuation (lambda (baz) (set! foo baz) 1))))) (bar) (+ 3 (foo (+ 1 1)))
16:47 rudybot: Staloras: ; Value: 3
16:47 Staloras: Okay, there we go.
16:47 Staloras: I'm clearly terrible at telling bots what to do.
16:47 Staloras: so... it looks like it did what I thought it would do
16:47 Staloras: probably.
16:52 Staloras: Thank you for the help, though.  Hopefully I'll get to the point where I can actually not fail at these things independently.
16:53 (quit) Staloras: Quit: Page closed
17:03 (quit) DanBurton: Quit: Leaving
17:10 (quit) gf3: Excess Flood
17:12 (join) gf3
17:46 asumu: 
18:01 (join) ernestas
18:04 (join) jrslepak_
18:54 (quit) jao: Remote host closed the connection
19:16 (join) dnolen
19:33 (nick) samth -> samth_away
19:53 (quit) mceier: Quit: leaving
20:01 (join) jao
20:09 (quit) dmac1: Ping timeout: 276 seconds
20:18 (quit) wishbone4: Remote host closed the connection
20:43 (join) dmac1
20:54 (quit) jonrafkind: Ping timeout: 260 seconds
21:03 (quit) jrslepak_: Quit: This computer has gone to sleep
21:04 (join) yin
21:13 (join) jonrafkind
21:21 (quit) mithos28: Quit: mithos28
21:37 (quit) dmac1: Ping timeout: 260 seconds
21:41 (quit) yin: Quit: Leaving
21:46 RacketCommitBot: [racket] plt pushed 2 new commits to master: http://git.io/u6jGCg
21:46 RacketCommitBot: [racket/master] add call to 'test-results' - Robby Findler
21:46 RacketCommitBot: [racket/master] swapped the backwards triangles - Robby Findler
21:53 (join) noelw
22:12 (quit) jonrafkind: Ping timeout: 245 seconds
22:31 (nick) elliottcable -> mi
22:31 (nick) mi -> elliottcable
22:40 (join) mithos28
22:48 (quit) masm: Quit: Leaving.
22:49 (join) masm
22:53 (quit) masm: Client Quit