00:17 (join) mheld
00:47 (quit) mwolfe: Remote host closed the connection
01:08 (quit) jonrafkind: Ping timeout: 240 seconds
03:40 (join) stis
05:07 (quit) stis: Ping timeout: 276 seconds
05:42 _p4bl0: mmm the doc for refresh doesn't seem to require any parameters. It's implemented at window<%> level, and semantically much more cleaner, thanks :-).
05:48 _p4bl0: :-/ I have much less blinking when repainting canvas using on-paint than refresh :-/
05:50 (quit) coldhead: Read error: Connection reset by peer
05:50 (join) coldhead
06:09 (join) mceier
06:10 (quit) coldhead: Quit: bababadalgharaghtakamminarronnkonnbronntonnerronntuonnthunntrovarrhounawnskawntoohoohoordenenthurnuk
06:46 (quit) mheld: Quit: mheld
07:02 (quit) mceier: Quit: leaving
07:05 (join) masm
07:59 (join) lucian_
08:01 (nick) lucian_ -> lucian
08:19 (join) MayDaniel
08:28 (quit) lucian: Remote host closed the connection
08:36 (quit) MayDaniel: Read error: Connection reset by peer
08:42 (quit) abbe: Changing host
08:42 (join) abbe
08:59 (join) corruptmemory
09:09 unkanon2: ok so I have a better example of a function that declares globals
09:10 unkanon2: what if, like sean grove does in his blog (but he does it in CL and I'm sure a macro would be fewer lines), we have a list of HTML tags that we want to turn into a function
09:11 unkanon2: e.g. '(a div img) will yield (define (a something) ("<a>~A</a>" something)) (define (div something) ("<div>~A</div>" something)) (define ... etc
09:27 (quit) unkanon2: Quit: Lost terminal
09:42 (join) unkanon-work
09:43 unkanon-work: oops, rebooted and forgot to reopen irssi
10:29 (join) MayDaniel
10:38 (join) lucian
10:59 (join) Nanakhiel
10:59 (join) mheld_
11:03 (quit) Lajla: Ping timeout: 264 seconds
11:03 (join) mheld
11:05 (quit) mheld_: Ping timeout: 240 seconds
11:09 (quit) mheld: Ping timeout: 246 seconds
11:15 (join) mheld
11:34 (join) sedeki
11:35 sedeki: hello all
11:36 unkanon-work: hi!
11:36 sedeki: hello
11:39 (quit) MayDaniel: Read error: Connection reset by peer
11:44 (join) anRch
11:46 (quit) lucian: Remote host closed the connection
11:53 askhader: ai
11:55 (quit) mheld: Ping timeout: 255 seconds
12:06 (join) jonrafkind
12:48 (join) mceier
12:51 sedeki: seriosuly, what could be wrong with this code...
12:51 sedeki: so frustrating
12:53 sedeki: now
12:57 sedeki: (make-array (shape 0 2) 1)
12:57 sedeki: how can i set different values on the array?
12:58 (quit) corruptmemory: Ping timeout: 240 seconds
12:58 (quit) tonyg: Remote host closed the connection
12:58 (quit) danking: Remote host closed the connection
13:01 (join) corruptmemory
13:02 (quit) anRch: Quit: anRch
13:03 jonrafkind: sedeki, you mean you want to mutate the array?
13:07 sedeki: i want to initilize it
13:07 sedeki: not just to one value
13:07 sedeki: i run several (array-set!) now
13:20 (nick) Nanakhiel -> Lajla
13:46 (join) carleastlund
13:52 (join) MayDaniel
14:00 (quit) MayDaniel: Read error: Connection reset by peer
14:14 carleastlund: clklein (or anyone who knows the answer), is there a way to change the way Redex's "test-results" displays values?  A little pretty-printing would go a long way.
14:16 clklein: carleastlund: I don't think so, but I'll double check.
14:18 carleastlund: clklein, does Redex use Rackunit or is it entirely separate?
14:20 clklein: carleastlund: It's entirely separate.
14:20 carleastlund: Hmm.  I have a large Redex model with lots of Greek letters in it.  I know I used Emacs, DrRacket, or both to write them by typing the LaTeX sequence, e.g. \alpha, then using some keybinding to automagically turn that into a UTF-8 character.  Anyone know what keybinding that is?  It has completely fled my memory.
14:21 clklein: It's esc-\ for me.
14:22 clklein: It's actually the individual test cases (e.g., with test-->, test-equal, etc.) that do the unweidly printing, not test-results.
14:22 carleastlund: In which?  And if in Emacs, in what mode?
14:23 clklein: These forms use print. I agree they should probably be pretty printing, but in the mean time, can you change the print handler? (I've never tried it. Maybe it's a pain.)
14:23 clklein: In DrRacket
14:24 carleastlund: Ah.  It's ctrl-\ on Mac.  Thanks, clklein and samth.
14:25 clklein: Oooh, duh. I'm on a Mac; I just don't know key names apparently :)
14:25 (quit) sedeki: Quit: L?mnar
14:25 carleastlund: Changing the print handler doesn't change the output.  Except that it is now followed by "#<void>".
14:26 clklein: hrm
14:26 clklein: which testing form are you using?
14:26 carleastlund: test-equal
14:29 clklein: It works for me. Be sure you're setting (current-error-port), not (current-output-port)
14:29 carleastlund: What does the port have to do with anything?
14:30 clklein: Don't you have to specify the port when you set the print handler?
14:30 carleastlund: current-print is a single parameter.  Are you referring to something else than that?
14:30 clklein: Oh, yea, I was using port-print-handler
14:31 carleastlund: Ah ha.  Then I should try that.
14:31 carleastlund: Sooooo much better, thank you.
14:32 clklein: I'll put changing the default on my TODO list.
14:35 carleastlund: clklein, can you remind me of the difference between #:attempts and #:retries in redex-check?  I can only find #:attempts in the documentation.
14:37 clklein: #:attempts says how many tests to run
14:38 clklein: #:retries say how many times to repeat generation of a `side-condition' pattern
14:38 clklein: #:retries should probably go away
14:39 clklein: I have NEVER used it, and people tend to be confused by it.
14:39 (join) stamourv
14:39 carleastlund: What does it default to?
14:39 clklein: 100
14:41 carleastlund: There definitely needs to be a way to control it.  Just better keyword names would help a lot.  Like "#:trials" and "#:side-condition-retries".  Those would clear up a lot.
14:41 clklein: Do you need to set it?
14:42 bremner: anyone using slideshow with a tiling window manager? slideshow doesn't seem to get along very well with xmonad; it still thinks it has the whole screen.
14:42 carleastlund: I have a side condition that is almost never satisfied -- there's really no way for me to find out if it is ever satisfied without scaling that number up.
14:43 clklein: I'm not sure that #:trials is any better than #:attempts, but I agree that #:side-condition-retries is better
14:43 clklein: oh, wait, it's actually slightly more general than side-condition patterns. Generation can also fail for _!_ patterns.
14:44 carleastlund: Don't call something an "attempt" if it's guaranteed to complete.  Redex-check doesn't "attempt" to test a property, it does test it.  That test may produce a negative result, but it is performed nonetheless.  Whereas random generation of side conditions *can* fail.
14:44 clklein: It attempted to find a bug
14:44 clklein: finding a bug can fail
14:44 carleastlund: If the macro were called "redex-find-bug", that'd be fine.
14:45 carleastlund: But it's not, it's redex-check.  It doesn't attempt to check.
14:46 clklein: I am partially convinced.
14:46 carleastlund: How about (redex-check foo #:check-retry 100 #:random-input-retry 2)
14:47 carleastlund: Rather than trying to differentiate them with generic words like "attempt", "trial", and "retry".  :)
14:47 clklein: That's better, but I don't think there's any name that's going to make "how many times to retry generation of any particular pattern" clear
14:48 clklein: It should almost be #:go-read-this-part-of-the-docs :)
14:49 carleastlund: Alternately, you could punt and allow customization via "#:input-generator <proc>", so people have to explicitly use generate-term to customize.
14:49 clklein: What would be proc's input?
14:49 carleastlund: I dunno.  Probably just a thunk.
14:50 clklein: How would it know what to generate?
14:50 carleastlund: Whatever generate-term's second form produces.
14:50 clklein: Wouldn't it need to take a pattern or somethign?
14:50 carleastlund: I am not a detail person.  :p
14:50 clklein: heh
14:51 carleastlund: Like, the thunk could just generate whatever, and redex-check would use its pattern to make sure it generated something good.  So if you used generate-term, there'd be a bit of redundancy.  It's not a perfect design... feel free to do something better than this.  :)
14:51 clklein: But anyway, I'm not sure that #:badly-named keyword is the best way for you to tell if the side-condition is every being satisfied
14:52 clklein: is it a side-condition on a reduction rule or a meta-function clause?
14:53 carleastlund: Neither, it's a side condition in a "define-extended-language".  It's used explicitly for the call to redex-check.
14:54 carleastlund: It's not something I really expect redex-check to be able to satisfy, but it's still important to be able to check many times in these cases to make sure.
14:56 clklein: so you set #:badly-named keyword to a big number to check and see if the exception goes away?
14:56 carleastlund: Yeah.
14:56 carleastlund: If, for instance, it turned out it was satisfied 1/200 of the time instead of 1/20 of the time, you'd want to leave that keyword at around 1000 instead of 100.
14:57 carleastlund: But really you almost want to set that in the grammar, because each side condition is going to behave differently.
14:57 clklein: yea, that makes sense
14:58 clklein: hrm
14:58 clklein: We've resisted polluting the Redex forms with random testing annotations so far.
14:59 clklein: Uh oh, I'm late for class. I'll look into fixing the name at the very least
15:28 (join) coldhead
15:36 (quit) coldhead: Read error: Connection reset by peer
15:36 (join) coldhead
15:44 (join) mheld
16:06 (quit) coldhead: Remote host closed the connection
16:12 (join) coldhead
16:50 (join) Samy
16:52 Samy: Can someone say to  me, is this Higher-order function http://codepad.org/3KNvBCZ5
16:52 Samy: pls
16:54 samth: Samy, no, that isn't higher-order
16:58 abbe: recursive function, that is.
17:04 (quit) tv|z: Ping timeout: 276 seconds
17:05 (join) tv|z
17:22 Samy: Is this higher order function ?  http://codepad.org/2tihMehw
17:24 (quit) Demosthenes: Ping timeout: 250 seconds
17:24 samth: Samy, no, it isn't
17:24 bremner: Samy: I don't think so,
17:24 samth: what are you trying to figure out?
17:26 Samy: why not ?
17:27 samth: `f' just adds its arguments
17:27 Samy: I must write a function sum all elements of a list. But this must be higher order function.
17:28 Samy: i mean  function "summe" must be a higher order
17:28 Samy: f not
17:29 (join) anRch
17:29 Samy: In my book i read, if a functon have another function as output, than it is a higher order function. this is what i try
17:31 (join) Demosthenes
17:33 jonrafkind: samth, isn't `ret' the end of a function?
17:34 samth: jonrafkind, return can occur multiple times in a function
17:34 samth: just like in C
17:34 jonrafkind: oh.. how are users supposed to know what size to give? just guess?
17:34 jonrafkind: i mean if the size is too big, will you get a segfault?
17:34 samth: yup
17:34 samth: sometimes
17:34 jonrafkind: :p
17:35 samth: sometimes it just disassembles more stuff
17:35 jonrafkind: oh wel, i guess its still mildly useful
17:35 samth: if you want to see what something does, you can get the info
17:35 jonrafkind: maybe matthew can add a size to the jit somewhere, like a nop
17:35 jonrafkind: what?
17:36 samth: i mean, you can try various values of size until you get what you want
17:36 jonrafkind: yea.. except a segfault makes it annoying
17:36 samth: mostly it doesn't segfault
17:36 jonrafkind: oh ok
17:36 samth: it only segfaults if that memory is mapped for your process
17:36 samth: isn't rather
17:36 jonrafkind: yar
17:37 samth: so usually you just get the disassembly for some other code
17:37 jonrafkind: i suppose differentiating the two functions isn't all that hard
17:37 jonrafkind: actually, can you tell when a new function starts?
17:38 jonrafkind: of course that assumes all jit traces are sequential in memory
17:38 samth: right
17:38 samth: i personally hope that matthew will just tell me - oh, get the size from here
17:38 jonrafkind: it seems like matthew could add an end of function instruction for all jit traces
17:38 jonrafkind: yea, or it could be part of the native closure
17:41 (quit) Samy: Ping timeout: 265 seconds
17:44 Lajla: jonrafkind, you could have told samy to use fold.
17:44 Lajla: But you were too late.
17:44 jonrafkind: fold
17:45 Lajla: Your heinous lack of attention have deprived this innocent soul from an answer.
17:45 Lajla: jonrafkind, gone.
17:45 Lajla: Forever, I was too late.
17:45 Lajla: Also, I thought I was the only person with such a fetish for short names that I considered 'ret' an acceptable name for a return statement/function
17:45 jonrafkind: samth, was here, he is my mirror image
17:47 samth: :P
17:51 (join) MayDaniel
18:36 (quit) anRch: Quit: anRch
18:42 bremner: hrm. Anyone running 5.0.2 can try "gracket  test.rkt --ps" with http://paste.debian.net/103855/ ?
18:43 (join) Yann3
18:43 (quit) MayDaniel: 
18:45 (join) Sky_
18:45 Sky_: Hi
18:46 Sky_: I saved my program as Interaction and now i cant view my code anymore... is there anyway to get my code back?
18:47 jonrafkind: can you be more specific
18:47 jonrafkind: as to what the problem is
18:47 bremner: well, the output is blank for me.
18:47 jonrafkind: sorry, i was talking to Sky_
18:47 bremner: ok
18:47 Lajla: No one wants to help poor bremner. =(
18:48 Sky_: When I open the .rkt file with DrRacket I cant see my code
18:48 jonrafkind: Sky_, you mean its just a white background?
18:48 Sky_: I only see what my Program should do
18:48 Sky_: yep
18:48 Sky_: it shows my 3 lines code, the output of my program
18:49 jonrafkind: bremner, fwiw, in racket 5.0.9 it brings up a gtk box that asks me to save the file
18:50 jonrafkind: hm, but there is no output in the resulting ps file
18:50 jonrafkind: or at leave evince wont load it
18:51 Yann3: Hi, how can I include a file that isn't a module?  I just want to split up a big file into logical sections.
18:51 jonrafkind: you can make it a module easily enough with (provide (all-defined-out))
18:53 Yann3: jonrafkind: but I want to use normal code (not only functions) in the files I want to include
18:55 bremner: jonrafkind: yeah, that sounds like my experience as well
18:56 Sky_: Hmm can I upload my .rkt file and you say me what I have to do to see my source code again? :S
18:56 (quit) jonrafkind: Ping timeout: 276 seconds
19:02 (quit) Sky_: Quit: Page closed
19:07 (quit) Fill: Ping timeout: 240 seconds
19:09 (join) Fill
19:10 (join) Sky_
19:11 (quit) Sky_: Client Quit
19:24 (join) Yann1
19:25 (quit) Yann3: Ping timeout: 276 seconds
19:34 (quit) mceier: Quit: leaving
19:46 (join) unkanon
19:47 (quit) masm: Quit: Leaving.
20:05 (join) masm
20:17 Yann1: Hi, how can I include a file that isn't a module?  I just want to split up a big file into logical sections.
20:19 Lajla: Ehh
20:19 Lajla: using (load ...) I guess
20:24 Yann1: Lajla: this don't work. I get ">gracket -t "myfile.rkt"
20:24 Yann1: Lajla: myloadfile.rkt:2:0: compile: unbound identifier (and no #%app syntax transformer is bound) at: #%top-interaction in: (#%top-interaction module functions scheme (#%module-begin (define (unshift list....[code]
20:25 Lajla: Yann1, are you using (load "/somepath/to/somefile.rkt") or something similar?
20:25 Yann1: Lajla: yes
20:27 Lajla: Then I have no idea.
20:28 bremner: Yann1: make a small test case, and if that doesn't work, paste your code to some pastebin site
20:28 Yann1: bremner: ok
20:33 Yann1: bremner: file1: http://pastebin.com/aHsHwAxk file2: http://pastebin.com/NNj98NCg
20:35 Yann1: bremner: this is the error I get: http://pastebin.com/G9YMrkg8
20:39 (quit) masm: Quit: Leaving.
20:45 (quit) carleastlund: Quit: carleastlund
21:02 (join) _danb_
21:27 (quit) Demosthenes: Ping timeout: 255 seconds
21:45 (join) Demosthenes
21:45 (quit) parcs: Ping timeout: 240 seconds
21:47 (join) parcs
23:06 (join) mwolfe
23:07 (quit) corruptmemory: Quit: Leaving