00:02 (join) jao
00:04 (join) espringe
00:04 espringe: In racket repl (or dr racket) is there a way to refer to the result of the last expression?
00:06 espringe: Ah sweet, found it in xrepl
00:07 asumu: espringe: Was just about to direct you there. :)
00:07 espringe: asumu: Is it possible to get this in drracket?
00:08 asumu: espringe: Not as far as I know, but maybe try (require xrepl) in the definitions area.
00:08 espringe: Doesn't work :(
00:11 asumu: Yeah, I can't get it to work either. You could try asking eli if it's possible.
00:11 espringe: Having a ^ in drracket would be nice. But I'll see if I can get xrepl to do indenting and paren highlighting, and then i could use that
00:12 asumu: espringe: If you run the Racket REPL in emacs you could probably get that "for free"
00:13 espringe: I've only got enough room in my brain for vi hotkeys. I tried the emacs tutorial once, but I couldn't get anything to stick
00:14 asumu: Also, readline can do some basic paren matching (e.g. "set blink-matching-paren on" in .inputrc)
00:14 mithos28: I cannot see a (good) reason why xrepl shouldn't work in the drracket repl
00:14 asumu: espringe: I'm actually a vim user too. I get by using this mode in emacs https://gitorious.org/evil/pages/Home
00:29 (quit) jao: Ping timeout: 248 seconds
00:41 RackN00b: Can anyone maybe help me determine why this is failing? (define (remove-col polyomino)   (cond     [(empty? polyomino) empty]     [else (apply remove-first polyomino)]))      (define (remove-first list)   (rest list))
00:41 RackN00b: eek.. maybe I'll pastebin that!
00:42 RackN00b: http://pastebin.com/gf6GmLAx
00:42 offby1: what does "fail" mean?
00:42 RackN00b: I'm trying to remove the first first element of each nested list... remove-first: expects only 1 argument, received 2
00:43 espringe: Try removing the apply?
00:43 espringe: like:
00:43 espringe: (remove-first polymonin)
00:43 espringe: ?
00:43 offby1: yep
00:43 RackN00b: I was hoping to avoid recursing... If I take out the apply, I'll only be removing the first sublist...
00:43 RackN00b: I need to remove the first element in each sublist.
00:44 espringe: is polyomino a list of lists?
00:44 RackN00b: yes.
00:44 espringe: Then the problem is with your definition of remove-first
00:44 RackN00b: hmmm
00:44 espringe: I don't know enough lisp to help you, :(
00:44 espringe: so maybe someone else can
00:44 RackN00b: No worries... good place to start looking :D
00:44 espringe: (i'm a super noob)
00:45 RackN00b: <------ ridiculous Noob
00:45 Pauan: I would but I'm busy in-game
00:45 RackN00b: hey no worries...
00:45 espringe: does:
00:45 espringe: (define (remove-first list)   (apply rest list))
00:45 espringe: does that work?
00:45 espringe: and remove the "apply" from remove-col
00:46 RackN00b: nope
00:46 espringe: http://pastebin.com/4gJQw4E5
00:46 espringe: That's my totally uneducated guess
00:47 RackN00b: rest: expects only 1 argument but receive 2...
00:47 RackN00b: odd
00:47 RackN00b: Maybe I have to curry rest... I'll give that a shot
00:47 espringe: Pastebin all the code, that you're using to test
00:48 espringe: and i'll see if i can get it to work
00:48 espringe: good learning for me
00:48 RackN00b: Cool cool.
00:48 RackN00b: There's no code aside from what I put up.... I'll toss a check-expect on it and repost for you
00:50 RackN00b: http://pastebin.com/etZxfDQp
00:54 RackN00b: Got it... (map rest polyomino)
00:58 RackN00b: Is there a straightforward way to apply a given function n number of times as opposed to writing an accumulator that counts from 0 to n?
00:58 (join) offby1_
00:59 espringe: RackN00b: Ah, I see you've solved it. Here was my version: http://pastebin.com/K8Q8Trd5
00:59 espringe: but yeah, same solution really
01:00 espringe: There might already be a version of "safe-rest" in the standard functions, i have no idea :D
01:00 RackN00b: I don't know if there is or not, but I managed to get rid of one extra helper so yay!
01:01 (quit) dnolen: Quit: dnolen
01:02 Pauan: RackN00b: If you want to apply a function n number of times, I guess you could use build-list, but there may be a better way in Racket.
01:02 Pauan: unfortunately that has the effect of returning a list
01:03 jonrafkind: rudybot, (for ([i (in-range 0 5)]) (printf "x"))
01:03 rudybot: jonrafkind: ; stdout: "xxxxx"
01:03 Pauan: oh, nice
01:04 Pauan: (I don't use for stuff in Racket, in case you couldn't tell)
01:04 jonrafkind: you're a map kind of guy eh
01:04 Pauan: I'm an Arc kinda guy
01:05 Pauan: you'd use (repeat 5 (pr "x")) in Arc
01:05 RackN00b: Sorry Pauan - should have been clearer... I need to apply a function to a list n number of times... So given my list of lists (all of length n) to add a column, I add an empty character at the front of each nested list while adding a row amounts to adding a new list of length n to the front of the outer list... I just need to take in a value as to how many times I need to do this in both cases...
01:06 (quit) jeapostrophe: Ping timeout: 260 seconds
01:06 Pauan: (in case you're wondering what I'm doing on the Racket IRC, well... Arc is implemented in Racket, and I'm writing my own Arc compiler in Racket, and was having some issues so I wound up here)
01:07 Pauan: (so I ended up learning a bunch of stuff about Racket as a means to an end)
01:07 espringe: Pauan: I actually just heard about arc today. Is it significantly different than racket?
01:07 Pauan: yes
01:07 jonrafkind: whats interesting about arc? it looks like it enjoys being terse
01:07 espringe: What do you think is a better choice for someone new to lisp?
01:08 Pauan: it really does depend on what you want to get out of it
01:08 espringe: I'm trying to implement some really old genetic algos in lisp
01:08 espringe: like the original papers
01:08 Pauan: jonrafkind: I like it because it's very short and convenient, and I think it made an awful lot of right choices for a lisp
01:09 jonrafkind: I am not impressed with arc's default choice for spelling behavior. 'def', 'fn', 'mac'
01:09 Pauan: but that's the best part :D
01:10 Pauan: I can barely stand a Lisp that uses "lambda" anymore
01:10 jonrafkind: oh boy.. we are on opposite ends of the spectrum
01:10 Pauan: apparently so
01:10 Pauan: Racket is significantly different from Arc, so that doesn't surprise me
01:10 jonrafkind: 'fn' is to 'lambda' as aol speak is to english
01:11 jonrafkind: i dont understand why engineers despise grammatical and spelling errors but are ok with doing it in code
01:12 asumu: I prefer λ
01:12 espringe: RackN00b: is this what you were after? http://pastebin.com/T8ajB0r8
01:13 Pauan: asumu: that's nice, yeah, but keyword support is pretty poor
01:13 Pauan: so you end up needing a keyboard shortcut
01:13 espringe: Like apply a function over and over?
01:13 espringe: Or have an editor that coverts \ to it
01:14 Pauan: right, or that
01:14 espringe: i don't mind haskell style
01:14 jonrafkind: Pauan, you would hate my language, honu, instead of 'lambda' I use 'function' !
01:14 espringe: which is that backslash, that editors can normally optionally render as a lambda sign
01:14 Pauan: jonrafkind: I've done enough JavaScript, thanks
01:14 RacketCommitBot: [racket] plt pushed 2 new commits to master: http://git.io/Txic9w
01:14 RacketCommitBot: [racket/master] update the mac os x icons to use the newer, higher-resolution images - Robby Findler
01:14 RacketCommitBot: [racket/master] fix bugs in the way the Racket|Disable Tests menu item worked - Robby Findler
01:15 jonrafkind: but of course since you are in racket you can redefine/import any name you wish
01:15 Pauan: sure
01:15 Pauan: but I think defaults matter
01:16 jonrafkind: ok aside from that "trivial" issue it looks like arc uses non-hygenic by default macros?
01:16 Pauan: yeah
01:17 jonrafkind: strange design choice
01:17 Pauan: I guess you could see Arc as being like a blend of Scheme and CL
01:18 Pauan: nil/t for #f/#t for instance
01:18 jonrafkind: some people just can't let the 1960's die
01:37 (quit) freakazoid: Quit: Computer has gone to sleep.
01:54 asumu: BTW: I'm glad SICP's alternative never caught on: "It would be clearer and less intimidating to people learning Lisp if a name more obvious than lambda, such as make-procedure, were used."
01:54 asumu: Though I agree with the lambda bit (which I didn't remember being there, but I haven't read this in a while).
01:56 jonrafkind: did they say the same thing about car/cdr?
02:05 asumu: No, but it does mention in a footnote what they stand for.
02:09 Pauan: where are the unit tests for Racket located?
02:13 jonrafkind: collets/tests/racket
02:13 Pauan: thank you
02:15 (quit) EmmanuelOga: Ping timeout: 244 seconds
02:18 Pauan: also, what's the preferred way to get a change into Racket?
02:18 Pauan: fork the project on github and send a pull request, or something else?
02:18 jonrafkind: I think pull requests work
02:19 mithos28: Pauan: github pull requests usually work well, thats what I use
02:19 Pauan: okay, cool
02:22 (quit) realitygrill: Quit: realitygrill
03:03 Pauan: okay, it's officially Gmail that is mangling my e-mails
03:03 Pauan: at least this time the mangling is still readable for the most part
03:10 (quit) jonrafkind: Read error: Operation timed out
03:17 RackN00b: How would you guys go about counting the number of times a given number turns up in a nested list?
03:18 Pauan: hm
03:18 Pauan: does "count" work in a nested list?
03:19 RackN00b: I don't have "count" at my disposal... Might be limited to recursing through each sublist and adding to a running tally.
03:19 Pauan: nope it doesn't
03:19 Pauan: though you could flatten the list first :P
03:19 RackN00b: True...
03:24 Pauan: ozzloy: got the basic escapes working, but they seem to only work in pregexp mode.
03:27 (quit) noam: Read error: Connection timed out
03:28 (join) noam
03:31 (join) Blkt
03:44 Pauan: Also, I'm not entirely sure how to go about adding in \x and \u escapes.
03:44 Pauan: (let alone \X)
03:59 (join) Shvillr
04:02 (quit) 16SABNLGU: Ping timeout: 260 seconds
04:04 (quit) SidH_: Remote host closed the connection
04:04 (join) SidH_
04:28 (join) ahinki
04:43 (quit) mithos28: Quit: mithos28
05:04 (quit) SidH_: Remote host closed the connection
05:04 (join) SidH_
05:10 Blkt: good morning everyone
05:15 (join) tim-brown
05:28 ohwow: hi
05:32 (quit) ahinki: Remote host closed the connection
05:43 (quit) RackN00b: Quit: Page closed
05:59 (join) leandrosansilva
06:13 (join) Lajla
06:18 (join) noisetonepause
06:19 noisetonepause: hello
06:20 noisetonepause: if i've got a struct with ten fields and a list with the fields in the right order, is there a quick way to convert the list to the instance of the struct?
06:39 (join) epsil
06:40 epsil: if I want to add an element to the beginning of a list, I can do (cons element list)
06:40 epsil: but if I want to add it at the end, do I have to do (append list (list element)) ?
06:40 epsil: or is there an easier way
06:42 tim-brown: is there a standard scheme/racket indenting scheme (you know... the kind that is embraced by all users without complaint or controversy)
06:42 tim-brown: ?
06:49 epsil: tim-brown: I wasn't aware there was controversy
06:49 epsil: I let emacs indent the code for me
06:51 tim-brown: i ask for a standard, i expect controversy
06:57 epsil: I think the general idea in any lisp is that you should never attempt to indent it manually
06:57 epsil: always let the auto-indentation to the job for you (and do not fight it, even if you think you know better)
06:57 tim-brown: so where are the automatic rules described?
06:58 (part) noisetonepause
06:58 epsil: hm ... good question
06:58 epsil: the basic approach is very simple (line up parenthetical expressions with each other), but there are some exceptions, like define
07:05 epsil: *parenthetical expressions -> S-expressions
07:05 epsil: I know that in some lisps, you can define custom indentation rules for any form, overriding the standard rule
07:06 epsil: i.e., you would specify the indentation for a macro inside the macro definition
07:08 epsil: the code itself describes how it is to be indented
07:09 epsil: however, I doubt emacs goes to the trouble of looking up the indentation rules for basic things like define
07:10 epsil: that is likely hard-coded
07:10 tim-brown thought that emacs was even capable of psychoanalysing itself
07:11 (join) fridim_
07:12 epsil: well, elisp macros have indentation specifications
07:12 epsil: but I don't know if scheme macros have them
07:14 (quit) loz`: Read error: Connection reset by peer
07:15 (join) loz`
07:22 (quit) loz`: Read error: Connection reset by peer
07:22 (join) loz`
07:28 (join) jeapostrophe
07:30 (quit) loz`: Read error: Connection reset by peer
07:33 (join) loz`
07:39 (quit) loz`: Read error: Connection reset by peer
07:40 (join) loz`
07:58 (join) veer
08:11 (quit) leandrosansilva: Remote host closed the connection
08:18 (join) ahinki
08:20 (quit) epsil: Quit: WeeChat 0.3.5
08:34 (quit) veer: Quit: Leaving
08:52 (quit) noam: Read error: Connection reset by peer
08:52 (join) noam
09:32 (join) dnolen
09:39 (join) epsil
09:39 (quit) jeapostrophe: Ping timeout: 252 seconds
09:43 asumu: tim-brown: At least in Racket, generally how DrRacket indents is "correct" but that's also not universally preferred.
09:43 asumu: Taylor Campbell has a style guide that talks about indentation http://mumble.net/~campbell/scheme/style.txt
09:44 tim-brown: will peruse that
09:44 tim-brown: asumu: thx
09:53 (join) jeapostrophe
09:59 (join) kenjin_
09:59 (quit) jeapostrophe: Read error: Connection reset by peer
09:59 (nick) samth_away -> samth
09:59 kenjin_: ...
10:00 kenjin_: Hello, I have a question about placeholder in db library
10:01 kenjin_: (query-exec c "drop table if exists $1" 'foo) doesn't work
10:02 kenjin_: query-exec: SQL syntax error in "drop table if exists $1"
10:03 (quit) dnolen: Quit: dnolen
10:24 (part) shofetim: "ERC Version 5.3 (IRC client for Emacs)"
10:38 (join) freakazoid
10:46 (join) realitygrill
10:47 samth: kenjin_: i don't know much about sql, but might that be a syntax that's not supported by your particular db?
10:49 kenjin_: Now I understand why. It's because placeholders can only be placed where "values" are
10:50 kenjin_: BTW, I don't know whether you remember me, but I've succeeded in writing the syntactic sugar.
10:51 kenjin_: Now I realize syntax-case is easy to use really.
10:53 (quit) realitygrill: Ping timeout: 248 seconds
10:59 (quit) ahinki: Quit: ChatZilla 0.9.87 [Firefox 9.0/20111122192043]
11:23 (join) dnolen
11:32 (join) keenbug
11:38 (quit) freakazoid: Quit: Computer has gone to sleep.
11:38 (join) gumbo_
11:39 gumbo_: is there any plan to support html5 in racket's standard libraries?
11:45 (quit) gumbo_: Remote host closed the connection
11:46 (join) gumbo
11:46 samth: gumbo: what do you mean by "support html5"
11:46 samth: you can already write html5 web pages using the xml library, for example
11:47 gumbo: Thanks, I'm new to racket(not lisp), and when looking through the tutorials I wondered this :)
11:51 samth: gumbo: the real question is what you mean by "html5"
12:01 gumbo: HTML5+CSS3+JS
12:02 samth: so, racket provides quite nice support for constructing html documents
12:03 samth: racket doesn't have a standard library for constructing css files
12:16 (quit) epsil: Quit: WeeChat 0.3.5
12:23 (quit) Blkt: Quit: ERC Version 5.3 (IRC client for Emacs)
12:25 (join) jonrafkind
12:25 (join) masm
12:38 (join) MayDaniel
12:39 (join) mithos28
12:41 (quit) tim-brown: Quit: Leaving
12:54 (quit) mithos28: Quit: mithos28
13:02 (join) Guthur
13:02 Guthur: will DrRacket run on an ARM based linux distro?
13:05 jonrafkind: yea but theres no jit
13:05 (join) freakazoid
13:12 freakazoid: Has anyone built a SchMUSE clone with Racket?
13:17 samth: what is schmuse?
13:17 (join) carleastlund
13:27 Guthur: jonrafkind: and what is the up shot of no JIT?
13:28 jonrafkind: it will run but slowly
13:29 Guthur: ok, that shouldn't be a problem, it's for educational purposes
13:29 Guthur: jonrafkind: cheers
13:30 jonrafkind: ok, let us know if you have any problems. im not sure how well tested the arm version is
13:31 Guthur: ok will do, the project is still in the incubation stage so will be at least a month before a proper test
13:32 freakazoid: samth: Scheme MUSE
13:33 Guthur: looking at running it on this when the hardware is available -> http://www.raspberrypi.org/
13:33 Guthur: hopefully will be available next month, I'll pick one up then to test things
13:33 jonrafkind: with 256mb of ram it should work ok
13:47 (join) ed-k
14:22 (join) awest
14:23 awest: Have there been any reports of 5.2 bugs in the Mac version of DrRacket, regarding the prefixing of identifiers provided by a module?
14:25 (part) ed-k: "ERC Version 5.3 (IRC client for Emacs)"
14:28 jonrafkind: awest, not that I know of. whats the bug?
14:29 (quit) freakazoid: Quit: Computer has gone to sleep.
14:30 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/hjSTiw
14:30 RacketCommitBot: [racket/master] adjust drracket to clean up the code that sets the initial value of - Robby Findler
14:31 (join) freakazoid
14:32 (quit) keenbug: Ping timeout: 245 seconds
14:36 awest: jonrafkind: let me reproduce it a third time and then I'll describe/report it today or tomorrow.
14:49 (quit) kenjin_: Ping timeout: 252 seconds
14:50 (join) EmmanuelOga
14:51 (join) ahinki
14:52 (join) mithos28
14:58 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/iI7biQ
14:58 RacketCommitBot: [racket/master] set the current-load-relative-directory to #f (initially) in drracket - Robby Findler
15:25 (quit) awest: Quit: leaving
15:36 RacketCommitBot: [racket] plt pushed 2 new commits to master: http://git.io/rGpkMg
15:36 RacketCommitBot: [racket/master] add the ability to pass a bitmap directly for the splash screen - Robby Findler
15:36 RacketCommitBot: [racket/master] adjust the dock icon (under mac os x) to change on the weekends - Robby Findler
15:40 (join) epsil
15:57 (quit) ahinki: Remote host closed the connection
15:59 RacketCommitBot: [racket] plt pushed 3 new commits to master: http://git.io/efDu4Q
15:59 RacketCommitBot: [racket/master] make `in-sequences' and `in-cycle' accept 0 sequences - Matthew Flatt
15:59 RacketCommitBot: [racket/master] raco setup: one more path simplifcation point - Matthew Flatt
15:59 RacketCommitBot: [racket/master] ffi/unsafe/objc: at least load on platforms without libobjc - Matthew Flatt
16:19 (join) keenbug
16:22 (quit) keenbug: Read error: Operation timed out
16:34 (join) keenbug
16:55 RacketCommitBot: [racket] plt pushed 3 new commits to master: http://git.io/p6ILMA
16:55 RacketCommitBot: [racket/master] Adjusting the weekend-bitmap-swapping code so it changes dynamically - Robby Findler
16:55 RacketCommitBot: [racket/master] adjust the autosave error code to print the stacktrace - Robby Findler
16:55 RacketCommitBot: [racket/master] change file-creator-and-type so that it raises exn:fail:filesystem? - Robby Findler
17:00 (quit) epsil: Quit: WeeChat 0.3.5
17:00 (nick) Guest12546 -> cipher
17:00 (quit) cipher: Changing host
17:00 (join) cipher
17:14 (quit) freakazoid: Quit: Computer has gone to sleep.
17:14 (join) jao
17:21 (quit) noam: Read error: Connection reset by peer
17:21 (join) noam
17:33 (nick) samth -> samth_away
17:43 (quit) keenbug: Ping timeout: 244 seconds
17:43 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/VOxWHA
17:43 RacketCommitBot: [racket/master] on-receive does not have to exist for a world to be a part of a universe, weird but feasible; Closes PR12401 - Matthias Felleisen
17:49 (quit) gumbo: Read error: Connection reset by peer
18:02 (quit) dnolen: Quit: Page closed
18:03 (join) freakazoid
18:49 (quit) MayDaniel: Read error: Connection reset by peer
18:57 (quit) asumu: Quit: leaving
19:08 (quit) fridim_: Ping timeout: 248 seconds
19:23 (quit) carleastlund: Read error: Connection reset by peer
19:23 (join) carleastlund_
19:40 (quit) Guthur: Remote host closed the connection
19:44 espringe: This is what i'd consider to be the ultimate web programming, if i could do something like this:
19:44 espringe: http://pastebin.com/mXnDrSyd
19:44 espringe: And the output of that, would be an HTML page -- with a version of "make-bg-red" compiled into some javascript
19:44 espringe: Is anything like that, remotely possible?
20:05 jonrafkind: I suppose thats possible, you could make a pass over the code and convert racket code into javascript
20:06 jonrafkind: but you would have to take care to preserve the semantics of javascript
20:06 (join) realitygrill
20:20 Pauan: espringe: As jonrafkind said, that would require a Racket->JS compiler, which already exists: http://hashcollision.org/whalesong/
20:20 (join) asumu
20:20 jonrafkind: well thats for client side stuff
20:20 jonrafkind: i think..
20:20 Pauan: I'm assuming they wanted to run the JS on the client side
20:21 Pauan: so unless whalesong has some limitation, I don't see why you couldn't do the Racket->JS compilation on the server
20:21 Pauan: and then send it to the client
20:21 jonrafkind: i mean espringe wants to write a single piece of code that is automatically split between server side code and client side code
20:21 espringe: Yup
20:21 jonrafkind: I got the feeling that whalesong was purely lcient side
20:21 espringe: I want the server to compile the js, and send it to the client
20:21 jonrafkind: as in convert an entire racket program to js
20:21 espringe: I only want to convert a few functions to js
20:21 jonrafkind: with some restrictions then what espringe want's to do should be pretty easy
20:22 jonrafkind: it gets hard as you scale it up to the full racket language
20:22 Pauan: of course
20:22 espringe: I shouldn't need stuff like an FFI and what not
20:22 Pauan: because of limitations in JS or whalesong
20:22 espringe: But it should be reasonably complete
20:22 espringe: I might have a stab at it, it'd be pretty awesome
20:22 Pauan: but as long as you stick to stuff that's a reasonable mapping between JS and Racket, it should be easy
20:22 Pauan: (i.e. don't try anything crazy)
20:23 espringe: I'd really like totally arbitrary functions to work. E.g. if i wrote/used a (purely functional) md5 function, it should just work in both
20:23 espringe: and i can reuse the same code, from the client and server
20:23 espringe: It's a bit useless if i have to make special versions for js and special versions for the server
20:23 espringe: Like one cool thing would be form validation, and using the exact same methods on the client and server to validate
20:24 espringe: (client for convience, server for security)
20:24 Pauan: that'll depend on what the code does and whether whalesong can handle it
20:24 Pauan: no way to know for sure unless you try it
20:25 (join) RackN00b
20:26 espringe: The other thing it would be really cool for, is making "desktop" applications with browser-based frontend. That way you'd get really great portability between different OS's (including mobile ones) as well as running on an actual web server
20:26 Pauan: like with say Adobe AIR?
20:27 espringe: Sure, i guess. I've never used adobe air.
20:27 espringe: Does it compile down to just html?
20:27 espringe: (and js)
20:27 freakazoid: espringe: it has its own runtime
20:27 Pauan: Adobe AIR is a runtime for executing Flash, Flex, HTML, and Ajax on the desktop
20:27 freakazoid: which constantly uses 5% CPU whenever it's running
20:27 Pauan: Mozilla Prism is an alternative
20:27 freakazoid: so don't run it on a laptop
20:28 Pauan: http://en.wikipedia.org/wiki/Mozilla_Prism
20:28 espringe: The one thing I'd be missing is proper sockets. I like writing networked code :(
20:28 Pauan: (though it seems Prism has been discontinued...)
20:29 (join) francisl
20:32 (quit) jonrafkind: Ping timeout: 258 seconds
20:39 RackN00b: Hey guys - http://pastebin.com/eeE1PJVB
20:39 RackN00b: I was wondering if someone could help me troubleshoot this....
20:39 Pauan: can you give an example?
20:39 Pauan: for instance
20:39 Pauan: the input, and expected output?
20:40 RackN00b: Specifically, I am having trouble with a part of the code where I am trying to add a character to the front of each sublist.... Sure...
20:41 RackN00b: (check-expect (superimpose '((#\. #\. #\.)(#\X #\X #\X)) '((#\T #\T #\T)(#\T #\. #\T)) (make-pos 1 1)) '((#\. #\. #\.)(#\T #\T #\T)))
20:41 RackN00b: Does that make any sense ?
20:42 Pauan: yeah
20:42 RackN00b: I'll try and give you an example for just the blank-cols
20:42 Pauan: wait, where's make-pos?
20:45 (join) dnolen
20:51 RackN00b: Ah.. sorry , it's defined previously as (define-struct pos (x y))
20:51 Pauan: kay
20:51 Pauan: ah yes, the infamous build-2dlist
20:52 RackN00b: Would this involve currying the function blank-char?
20:52 RackN00b: Haha... yep Do you want me to post that code as well ? I know that part works so I can also comment it out...
20:52 Pauan: well it'd save me the trouble of digging through the logs, yeah
20:53 RackN00b: ok.. two secs and I'll pastebin it.
20:57 RackN00b: http://pastebin.com/1DAeLMpu
20:58 RackN00b: Now the rest of the function isn't yet written.. for the actual superposition of the top onto the bottom - I had a clunky function for it but I'd like to rewrite it more succintly with nice locals.
20:58 Pauan: by the way
20:58 Pauan: my suggestion would be to use "let" rather than "locals"
20:58 Pauan: local*
20:59 RackN00b: I haven't let before so I'll have to read up on it.
20:59 Pauan: (let ([a 5]) a)
20:59 Pauan: binds the local variable "a" to 5
21:00 Pauan: so (let ([a 5] [b 10]) ...) binds the local variable a to 5 and b to 10
21:00 RackN00b: Can the local variable be bound to a function?
21:00 Pauan: sure
21:00 Pauan: but in that case you can just use "define"
21:00 Pauan: as in...
21:00 Pauan: (lambda () (define foo ...) ...)
21:00 Pauan: a define inside a function is bound locally
21:00 Pauan: no need for local
21:01 Pauan: works for ordinary vars too, I believe
21:01 Pauan: not just local functions
21:01 RackN00b: Neat
21:01 Pauan: then there's the "let loop" form
21:01 Pauan: (let loop ((x 0)) ...)
21:01 Pauan: which creates a local function called "loop" and immediately executes it
21:01 Pauan: with x defaulting to 0
21:02 Pauan: this is convenient for writing recursive functions
21:02 Pauan: oh, in this case you want let* though
21:03 Pauan: In the code (let ([a 5] [b a]) ...) the variable "b" refers to the *global* value of "a"
21:03 RackN00b: Oh, that's nice.
21:03 Pauan: but in (let* ([a 5] [b a]) ...) the variable "b" refers to the local variable "a"
21:03 Pauan: and thus is 5
21:03 Pauan: in other words, let* is sequential
21:03 RackN00b: I can see how that'd be handy
21:03 Pauan: local variables can refer to previous local variables
21:03 Pauan: okay, first issue
21:04 Pauan: procedure blank-char: expects 1 argument, given 3: '((#\.)) '(#\T #\T) '(#\T #\.)
21:04 Pauan: you're *applying* blank-char to a list
21:04 Pauan: so first off, remove the apply
21:04 Pauan: and now I got '(#\. ((#\.)) (#\T #\T) (#\T #\.))
21:04 RackN00b: right
21:05 Pauan: hm...
21:05 Pauan: what is blank-rows supposed to be?
21:05 Pauan: it's ((.)) for me
21:05 RackN00b: Hmmm I'm getting something different.... Is that using the newest paste - i edited the definitions for top and bottom
21:06 Pauan: it's using this:
21:06 Pauan: http://pastebin.com/1DAeLMpu
21:06 Pauan: lemme try it with your version of superimpose
21:06 Pauan: same thing
21:06 RackN00b: I'm getting this: (list (list #\. (list #\.)) (list #\. #\T #\T) (list #\. #\T #\.))
21:07 RackN00b: Oh, and replaces apply with map, should have mentioned that...
21:07 Pauan: ah
21:07 Pauan: yeah, now I get the same thing
21:08 Pauan: why does top have only 2x2 elements, while bottom has 2x3?
21:08 RackN00b: I don't know... This is strange - it was working when I only had blank-rows - i fear I might have changed something small without knowing :O
21:09 Pauan: hm...
21:10 Pauan: okay, hold on, you're trying to superimpose bottom on top of top?
21:10 Pauan: so that the end result is that any "blank" characters in bottom end up replaced with those in top?
21:11 RackN00b: Yes but also non-blank characters... I am superimposing top onto bottom
21:11 RackN00b: or trying to at least :P
21:11 Pauan: oh
21:11 Pauan: hm...
21:11 (join) didi
21:12 Pauan: I'm not understanding the desired result in check-expect then
21:12 RackN00b: The reason I am writing this portion is to "pad" the top such that it is the same size as the bottom and I can overlay one onto the other with a simple function....
21:12 Pauan: Imposing top onto bottom, I'd expect this: ((#\T #\T #\B) (#\T #\. #\B))
21:13 Pauan: and imposing bottom onto top, I'd expect this: ((#\B #\B #\B) (#\T #\. #\B))
21:13 RackN00b: Sorry, the first is correct...
21:13 Pauan: okay, then
21:13 Pauan: so...
21:13 Pauan: first off, why do you have to pad it?
21:14 Pauan: why not just have top be 2x3?
21:14 RackN00b: Because the function I had to overlay the top onto the bottom consumes both lists stepwise so they have to be the same size.. I couldn't figure out how to do it without padding... Also, I coudln't figure out how to reference where to begin the overlay given the offset.
21:15 Pauan: well, why not change it so when it reaches the end of the list, it assumes a blank character?
21:15 Pauan: that way, if one list is shorter than the other, it's automatically padded?
21:16 RackN00b: But the padding also serves the purpose of establishing the correct placement given the offset... The first of the first of each list is (0 0) so if my offset is (1 1) I have to start one down and one over for my overlay... Can I accomplish that without padding?
21:17 RackN00b: brb... daughter wnat a bedtime snack! two mins... sorry!
21:17 Pauan: let's first get superimposition working, then we'll worry about offsets
21:22 RackN00b: sorry... kid tucked in :D
21:23 Pauan: okay
21:23 Pauan: got superimpose working
21:23 Pauan: it's kinda long, though
21:24 RackN00b: This is my overlay function I wrote...http://pastebin.com/Rgc9iiKq
21:24 Pauan: right, see, that'd be perfect for a let loop
21:25 RackN00b: I wish i could use them :( I mean it'd not horrible on something like this but I can certainly imagine it would be a lot more readable using let loops
21:25 Pauan: http://pastebin.com/YsRd7ENj
21:25 Pauan: expires in an hour
21:26 Pauan: see what I mean about being kinda long?
21:26 Pauan: it works properly with shorter lists, though
21:26 Pauan: as in, if top is shorter than bottom, it'll get padded automatically
21:26 Pauan: now to test it the other way around...
21:26 Pauan: ah, yeah, that breaks
21:26 RackN00b: I don't think I will ever have that situation though...
21:27 RackN00b: The bottom represents a partially solved puzzle while the top is simply a piece.
21:27 Pauan: so the bottom can never be smaller than the top?
21:28 RackN00b: Well if I pad the top to get the proper offsets then yes, but without padding, no.
21:28 Pauan: then my function should work
21:28 Pauan: okay then
21:28 RackN00b: But even in the case of padding, I can always "trim" the top to match the size of the bottom...
21:28 Pauan: you said you wanted to offset it, right?
21:28 RackN00b: that's the goal, yep
21:28 Pauan: I think that should be a separate function
21:29 Pauan: hm...
21:29 Pauan: first thing, though
21:29 Pauan: with this: ((#\T #\T) (#\T #\.))
21:29 Pauan: which is the X axis?
21:29 Pauan: the (#\T #\T) part?
21:29 Pauan: or both sublists?
21:29 RackN00b: Each sublist has an x and a y component...
21:30 Pauan: okay
21:30 Pauan: hm...
21:30 Pauan: but that doesn't explain this: ((#\B #\B #\B) (#\. #\. #\B))
21:30 RackN00b: So that could be though of as a square with four smaller squares where there are all T but the bottom right one.
21:30 Pauan: which part of that is x and which is y?
21:30 RackN00b: That one would be an L-tetromino
21:30 Pauan: okay, so #\. is the empty one
21:30 Pauan: hm
21:30 RackN00b: yes...
21:30 Pauan: L-shaped
21:30 Pauan: okay, gotcha
21:31 Pauan: that means each sublist is Y
21:31 Pauan: and the entire list is X
21:31 RackN00b: I guess the coordinates are implied in the representation... does that make sense?
21:31 RackN00b: OK... I follow... nice
21:31 Pauan: okay then
21:32 Pauan: so that means if you want to offset it by 1 value to the X
21:32 (quit) masm: Quit: Leaving.
21:32 Pauan: that involves adding a #\. to each sublist
21:32 RackN00b: You're adding a #\. to the ... beat me to it!
21:32 RackN00b: yes
21:32 Pauan: and if you want to offset it by Y
21:32 Pauan: then that involves adding a new list
21:32 Pauan: to the front of the list
21:32 RackN00b: add a list of #\. the same length
21:32 Pauan: well
21:32 Pauan: I don't think it strictly needs to be the same length
21:32 Pauan: if our automatic-padding code works
21:32 Pauan: point is it just needs to be something that's recognized as an offset
21:32 Pauan: okay then
21:33 Pauan: I'll whip up an offset function
21:33 RackN00b: ok...
21:33 Pauan: so rather than passing the offset to superimpose
21:33 Pauan: you'd first call the offset function on the list
21:33 Pauan: and then pass the offset list to superimpose
21:33 Pauan: two separate functions
21:33 Pauan: will that work?
21:34 RackN00b: It will as long as the offset is less than the difference between the width of the base and the top... otherwise, we'd be  superimposing larger onto smaller...
21:34 Pauan: in other words, rather than (superimpose bottom top (make-pos 1 1))
21:34 Pauan: you'd say this:
21:34 Pauan: (superimpose bottom (offset top 1 1))
21:34 Pauan: well, we could put in checks for that
21:34 Pauan: or define behavior: cutting off the excess
21:34 RackN00b: Sure... ok
21:34 Pauan: or growing the matrix to fit the larger
21:34 Pauan: etc.
21:34 Pauan: I'm not worried about that: let's get it working first
21:36 RackN00b: ok
21:38 Pauan: okay
21:38 Pauan: offset code works
21:38 Pauan: now let's try offsetting top by 1 and superimposing it onto bottom
21:40 Pauan: by the way
21:40 Pauan: personally, I would swap the args to superimpose
21:40 Pauan: so it's (superimpose top bottom)
21:40 Pauan: 'cause you're superimposing top onto bottom
21:41 RackN00b: agreed- it's misleading isn't it..
21:42 (quit) mithos28: Quit: mithos28
21:42 Pauan: okay
21:42 Pauan: so when you offset
21:43 Pauan: top might potentially be outside the range of bottom
21:43 RackN00b: Right, but we can trim it
21:43 Pauan: for instance, bottom only has two X coordinates
21:43 Pauan: okay
21:43 Pauan: so you want to trim it?
21:43 Pauan: to always be the size of bottom?
21:43 Pauan: an alternative would be to return a matrix that's the combination of both
21:43 RackN00b: I would think... we know that the offset is from (0 0)...... a matrix eh?
21:43 Pauan: well I'm thinking of them as matrixes
21:44 Pauan: easier to say than "list of lists of chars"
21:44 RackN00b: true
21:44 Pauan: anyways, I'm just asking about what the desired property is
21:44 Pauan: should it grow the matrix to fit both top and bottom inside?
21:44 Pauan: or cut it to the size of bottom?
21:45 RackN00b: Well the bottom is of constant size for a given puzzle so it wouldn't change during the execution of the solver so I would say it would be easier to cut it to size...
21:45 Pauan: okay then
21:45 Pauan: I'll make it cut it to the size of bottom
21:45 Pauan: that means that...
21:45 Pauan: offsetting top by 1x1
21:46 Pauan: and then superimposing it onto bottom...
21:46 Pauan: should be '((#\B #\B #\B) (#\. #\T #\T))
21:46 RackN00b: correct.
21:56 (quit) acarrico: Ping timeout: 255 seconds
21:56 Pauan: and it works
21:57 Pauan: it's big, hairy, clunky, but it works
21:57 RackN00b: wow.. you are good!
21:57 Pauan: now to pastebin it...
21:58 Pauan: http://pastebin.com/NBQQ5kHS
21:58 Pauan: expires in an hour, as usual
21:58 Pauan: oh, wait
21:58 Pauan: there's tons of unnecessary whitespace
21:58 Pauan: gimme a sec
21:58 RackN00b: That's ok...
21:59 Pauan: http://pastebin.com/Kybks35a
21:59 Pauan: done
21:59 Pauan: I really do think superimpose is too clunky, though
21:59 RackN00b: quick question regarding your use of currying... (curry cons #\.) ==> Am i correct in thinking that this adds the #\. to each sublist?
21:59 Pauan: but I'm not entirely sure how to make it better
21:59 Pauan: yes
22:00 RacketCommitBot: [racket] plt pushed 2 new commits to master: http://git.io/3IEOuw
22:00 RacketCommitBot: [racket/master] allow strings and symbols for name clause in world; Closes PR12403 - Matthias Felleisen
22:00 RacketCommitBot: [racket/master] doc bugs fixed; Closes PR12404 - Matthias Felleisen
22:00 Pauan: it's the equivalent of (lambda (x) (cons #\. x))
22:00 RackN00b: Cool..
22:00 Pauan: by the way
22:00 Pauan: to shift along the Y axis
22:00 Pauan: I just added the empty list
22:00 RackN00b: oh ok...
22:00 Pauan: so (offset top 1 1) returns '(() (#\. #\T #\T) (#\. #\T #\.))
22:00 Pauan: but the superimpose code can handle that okay
22:01 Pauan: hrm... I really don't like that superimpose code, though
22:03 RackN00b: I think it's pretty sweet. In any case, I have to run - all-nighter last night and am I barely managing to stay awake tonight. I'll have to study the functions tomorrow morning and see what I can learn from them... Thanks so much Pauan - you've been a huge help in helping me understand.
22:03 Pauan: sure
22:04 RackN00b: Hopefully one day when I get to be mroe of a guru I can pay it forward! Cheers for now!
22:04 (part) RackN00b
22:08 (join) acarrico
22:16 (join) delleps
22:17 RacketCommitBot: [racket] plt pushed 2 new commits to master: http://git.io/lUPFIA
22:17 RacketCommitBot: [racket/master] disable file in drdr - Robby Findler
22:17 RacketCommitBot: [racket/master] fix a bug in the module lexer; it was returning the wrong length for the tokens - Robby Findler
22:17 (quit) didi: Quit: ERC Version 5.3 (IRC client for Emacs)
22:21 (quit) delleps: Ping timeout: 265 seconds
22:46 (join) mithos28
22:47 (quit) freakazoid: Quit: Computer has gone to sleep.
22:54 ozzloy: Pauan, i expected the section would be pregexp and not regexp
22:54 ozzloy: Pauan, cool that you made some progress though.
22:54 Pauan: any ideas about how to handle \u and \x though?
22:55 ozzloy: Pauan, i am torn.  i'd like to work on that, but i also want to go through the ai-class.com homework before wednesday
22:55 ozzloy: Pauan, also consider that adding to regexp is more likely to break someone's code somewhere that relies on \\n actually matching "n"
22:56 ozzloy: whereas pregexp throws a handy error
23:05 (quit) carleastlund_: Quit: carleastlund_
23:05 Pauan: yeah, I know
23:05 ozzloy: Pauan, i'd start by looking at how different other parts of regexpes are handled
23:05 Pauan: I'm okay with it being for pregexps
23:05 Pauan: it's just annoying because regexp-match (and friends) make a regexp if you give them a string
23:05 Pauan: rather than a pregexp
23:05 ozzloy: like stuff between the (? and the : and stuff between a [ and a ]
23:06 ozzloy: how [:alnum:] is handled
23:06 ozzloy: to see how the code currently consumes a token
23:06 Pauan: right, I figured I'd need to look at strings.c too
23:06 Pauan: to see how Unicode is handled
23:06 ozzloy: oh good call
23:07 ozzloy: pun intended
23:15 (join) freakazoid
23:26 (join) jonrafkind
23:27 (quit) dnolen: Quit: dnolen
23:48 (join) dnolen