00:03 Nisstyre: verahsa: http://ideone.com/5E1vF9
00:03 Nisstyre: that's the equivalent match version of sublist
00:04 Nisstyre: er, I messed up the subtraction part
00:04 Nisstyre: let me fix that
00:05 Nisstyre: http://ideone.com/G4p3Jl
00:05 verahsa: That's returning false on ones that should be true
00:06 verahsa: the sublist function is doing precisely what it's supposed to, digging through and finding if the search list is contained anywhere within the list to be searched
00:06 verahsa: the one i have, i mean
00:06 (quit) kvda: Quit: z____z
00:08 Nisstyre: yeah I know I'm just showing how you might do it with match
00:09 verahsa: ...
00:09 verahsa: and yet the version with match that you showed me doesn't actually do that, unless i'm blind somehow
00:09 asumu: ksefchik: what are you seeing?
00:10 ksefchik: DrRacket seems to be crashing when I close the window of the animate function in universe.ss
00:10 ksefchik: But it seems random. Doesn't crash all the time.
00:11 asumu: ksefchik: which version of Racket?
00:11 ksefchik: 5.3.3 Win x64
00:12 Nisstyre: verahsa: yeah I made a small mistake
00:12 Nisstyre: give me a minute
00:14 Nisstyre: verahsa: http://ideone.com/K7I8h9
00:14 Nisstyre: that should work
00:14 Nisstyre: anyway, that's how you use match
00:14 asumu: ksefchik: that's odd and shouldn't happen. Can you submit this as a bug report? (there's a menu item for it in DrRacket)
00:14 asumu: ksefchik: also any other info about the crashes would be helpful (include in bug report please)
00:15 ksefchik: Sure thing!
00:16 Nisstyre: verahsa: list-no-order should allow you to do what you want with the lgrep procedure
00:17 verahsa: Nisstyre: Thanks! :) However... um... is there a way to run two commands under an if with no else of some nature?
00:20 Nisstyre: verahsa: in the condition part of it?
00:20 Nisstyre: not sure exactly what you mean
00:20 Nisstyre: but let me hack up the match version of lgrep
00:23 Nisstyre: verahsa: so lgrep is supposed to find all the occurences of a in b right?
00:23 ksefchik: asumu: Sent.
00:23 asumu: ksefchik: Thanks!
00:24 ksefchik: No problem. I gotta admit, I kinda dig this language. Keep up the great work!
00:24 verahsa: Nisstyre: Yes, and then display any list that *does* contain said occurrence
00:24 (quit) ambrosebs: Remote host closed the connection
00:25 Nisstyre: verahsa: but you want to return all of the positions of the occurrences right?
00:25 Nisstyre: or no?
00:25 verahsa: Nisstyre: http://ideone.com/irqZtX The last line of that is what should be returned for (lgrep SL1 lst)
00:25 Nisstyre: oh I see
00:26 Nisstyre: you can just use filter then
00:26 verahsa: *blink*
00:26 Nisstyre: have you not been introduced to that yet?
00:27 verahsa: No.... this is a project that I've been assigned to be exposed to Racket/Scheme. I've gotten everything I needed figured out except how to *both* run sublist *and* recursively run lgrep on a (b-1) if the size of b is not 0
00:28 verahsa: cond works, but once sublist returns true and prints that list, it quits
00:28 verahsa: cond works in the other direction, and will run lgrep until there's not enough left in b to check, but doesn't output anything because it simply stops when it finds a true expression
00:28 verahsa: hence why i was asking about running two commands at once. :)
00:29 Nisstyre: verahsa: filter will take a procedure that takes one argument, and go through a list and apply that procedure to each element, it expects a boolean return value from each one
00:29 Nisstyre: the ones that return false will be "filtered" out
00:29 Nisstyre: everything else gets returned in a list
00:29 Nisstyre: it's exactly what you want here
00:29 verahsa: or in this case a list of lists
00:30 Nisstyre: but if you need to do it using a plain procedure you can use letrec and write a recursive procedure that just does what filter does
00:31 Nisstyre: verahsa: the last line of your lgrep (using filter) would look like "else (filter ((curry sublist?) a) b)"
00:31 Nisstyre: I realize that looks kind of like magic :P
00:31 Nisstyre: but it works
00:31 Nisstyre: http://ideone.com/sBkMMg
00:32 Nisstyre: would you get in trouble if you used that?
00:32 Nisstyre: I guess it would be better if you fully understood the solution
00:34 verahsa: Honestly, if you could explain  the 'curry sublist?' bit, I think I'd be understanding what was going on
00:34 Nisstyre: okay, it's pretty simple
00:34 Nisstyre: all curry is doing is creating a function that returns a new function each time you give it an argument
00:35 Nisstyre: so like, if you "curried" + then (+ 2) would return a function that does (+ 2 a)
00:35 Nisstyre: that sort of thing
00:35 Nisstyre: it's useful when you want to "fix" a parameter, i.e. make it the same
00:36 Nisstyre: so like ((curry *) 2) is equal to (lambda (x) (* x 2))
00:36 Nisstyre: you can change it to a lambda if you want
00:36 verahsa: the "lambda" i've seen enough to vaguely understand. :)
00:36 Nisstyre: yeah
00:37 verahsa: I'll be honest, being exposed to racket over the past two weeks has been vastly interesting considering my other experiences have been visual basic and c++
00:37 verahsa: things i love about it, and a few things that are just downright confusing. *grin*
00:37 verahsa: well, for me anyways
00:37 Nisstyre: yeah it is very confusing when you first start with this style of programming
00:37 Nisstyre: but once you get used to it, I find it is much easier to do than the sort of style C++ prefers
00:38 verahsa: I can see either being easy, with advantages and disadvantages, but i've also only been learning to program in general for about 4 years seriously and about 18 years as the occasional hobby
00:38 Nisstyre: verahsa: what kind of course are you doing?
00:38 verahsa: this course specifically is Theory of Programming Languages
00:38 Nisstyre: ah cool
00:38 Nisstyre: is this an online one?
00:39 verahsa: They offer it online, but I'm actually attending, this is one of the open labs.
00:39 Nisstyre: sounds cool
00:39 Nisstyre: I know someone who took an intro CS course last fall that was all Racket
00:39 verahsa: I've been enjoying learning about *how* programming works on a detailed level
00:39 Nisstyre: that was at the university of Waterloo
00:40 Nisstyre: yeah it's very fascinating learning about programming language theory
00:40 verahsa: programming in college isn't about learning a language, it's about learning how to learn a language, because who the **** knows what language you'll be required to work with in a career. :D
00:40 Nisstyre: exactly
00:40 Nisstyre: also, I think it's important to not get hung up on a certain style of programming language
00:40 Nisstyre: or to label them as being one style
00:41 Nisstyre: each one has advantages and disadvantages
00:41 verahsa: *nod* more exposure is always a good thing... things from C helped me understand racket, and things from racket have made things in C that I wasn't completely sure of a little more clear
00:41 verahsa: and then someone tosses out curry and my brain segfaults. *snicker*
00:41 Nisstyre: I think your mind will be blown even more if you ever try Haskell
00:41 Nisstyre: all the functions are automatically curried in that language
00:42 verahsa: I *really really* appreciate all your help, and I'll note Haskell down as something to look in to. Thank you SO much. :D Check your private messages for a relevant private question, though, here in a sec.
00:42 Nisstyre: okay I see it
00:45 (part) verahsa
01:09 (join) ynniv
01:13 (quit) mceier: Quit: leaving
01:17 (quit) ynniv: Quit: ynniv
01:29 (join) ksefchik
01:41 (quit) ksefchik: Ping timeout: 245 seconds
01:47 (quit) dented42: Ping timeout: 250 seconds
01:52 (join) dented42
02:00 (quit) Fare: Ping timeout: 272 seconds
02:09 (quit) sstrickl: Read error: Connection reset by peer
02:09 (join) sstrickl
02:09 (quit) sstrickl: Changing host
02:09 (join) sstrickl
02:18 (join) mceier
02:34 (join) BakaKuna
02:39 (quit) eikonos: Quit: Leaving.
02:42 (join) kvda
02:42 (join) eikonos
02:42 (quit) eikonos: Changing host
02:42 (join) eikonos
02:44 (join) ambrosebs
02:48 (join) basdirks
02:49 (quit) ASau: Ping timeout: 250 seconds
02:51 (join) soegaard
02:56 (quit) kvda: Quit: z____z
03:03 (join) vipjun
03:03 (join) rohni
03:14 (join) jao
03:14 (quit) jao: Changing host
03:14 (join) jao
03:15 (quit) jonrafkind: Ping timeout: 245 seconds
03:15 (join) egnarts-ms
03:16 (quit) rohni: Quit: Leaving.
03:17 (quit) dsantiago: Ping timeout: 240 seconds
03:18 (quit) noam: Ping timeout: 255 seconds
03:18 (join) noam
03:29 (join) ivan\
03:35 (quit) ambrosebs: Remote host closed the connection
03:38 (quit) noam: Read error: Connection reset by peer
03:39 (join) noam
04:00 (join) mithos28
04:13 (quit) mithos28: Quit: mithos28
04:19 (join) bitonic
04:19 (quit) jao: Ping timeout: 276 seconds
04:26 (quit) soegaard: Quit: soegaard
04:30 (quit) cdidd: Remote host closed the connection
04:33 (quit) racycle: Quit: racycle
04:47 (join) cdidd
04:48 (join) rohni
04:50 (quit) cdidd: Remote host closed the connection
04:51 (join) codeblack
04:58 (part) tobih
04:59 (join) cdidd
05:05 (join) tobih
05:13 (part) tobih
05:14 (quit) vipjun: Ping timeout: 245 seconds
05:27 (quit) rohni: Quit: Leaving.
05:33 (join) Aune
05:43 (quit) Aune: Quit: Hath Deprated
05:43 (join) Aune
05:44 (join) tobih
05:49 (quit) cored: Ping timeout: 260 seconds
05:50 (join) cored
05:50 (quit) cored: Changing host
05:50 (join) cored
05:54 (join) bjz
05:56 (join) vipjun
06:09 (join) vkz
06:17 (quit) yacks: Ping timeout: 245 seconds
06:23 (join) Skola
06:24 Skola: hello, I'm trying to make a hash with (hash), but I'm not sure how to give it the k:v pairs
06:24 (join) ambrosebs
06:27 (quit) cored: Ping timeout: 276 seconds
06:28 (join) cored
06:28 (quit) cored: Changing host
06:28 (join) cored
06:31 (join) rohni
06:34 Skola: nvm
06:39 (quit) Kaylin: Read error: Connection reset by peer
07:11 (quit) vkz: Quit: vkz
07:21 (join) jeapostrophe
07:53 (join) didi`
07:53 (nick) didi` -> didi
07:56 (join) vkz
07:59 (quit) bjz: Ping timeout: 248 seconds
08:06 (join) yacks
08:35 (join) anonymous97890
08:42 (join) mizu_no_oto
08:44 (quit) anonymous97890: Quit: leaving
08:50 (join) zyoung
08:57 (quit) rohni: Quit: Leaving.
09:01 (join) serhart
09:09 (join) bjz
09:14 (join) rohni
09:14 (quit) bitonic: Read error: Operation timed out
09:16 (quit) mizu_no_oto: Quit: Computer has gone to sleep.
09:28 (join) Fare
09:35 (quit) rohni: Quit: Leaving.
09:43 (join) rohni
09:56 (quit) rohni: Quit: Leaving.
10:02 (join) mizu_no_oto
10:02 (join) rohni
10:03 (join) bitonic
10:10 (join) anRch
10:13 (quit) ambrosebs: Remote host closed the connection
10:14 (quit) Aune: Quit: Hath Deprated
10:29 (quit) BakaKuna: Quit: BakaKuna
10:32 (quit) bitonic: Ping timeout: 245 seconds
10:40 (quit) sstrickl: Quit: sstrickl
10:47 (join) bitonic
10:53 (join) ambrosebs
11:04 (quit) cdidd: Remote host closed the connection
11:13 (join) soegaard
11:14 (quit) bitonic: Ping timeout: 264 seconds
11:16 (quit) anRch: Quit: anRch
11:18 (join) sstrickl
11:21 (join) racycle
11:21 (quit) mceier: Quit: leaving
11:21 (join) didi
11:23 (join) anonymous97890
11:24 anonymous97890: rudybot: (define a -nan.0) (number? a) (positive? a)
11:24 rudybot: anonymous97890: You can get a NaN on i386/amd64 in earlier versions, but it's trickier; you have to know what you're doing, and set up the bits right.
11:24 (join) jbclements
11:26 stamourv: rudybot: eval (define a -nan.0) (number? a) (positive? a)
11:26 rudybot: stamourv: ; Value: #f
11:26 stamourv: rudybot: Is that it?
11:26 rudybot: stamourv: was going to voluenteer twb, but if you want me that badly then i cna't decline
11:26 stamourv: rudybot: eval (define a -nan.0) (values (number? a) (positive? a))
11:26 rudybot: stamourv: ; Value: #t
11:26 rudybot: stamourv: ; Value#2: #f
11:26 stamourv: rudybot: That's better.
11:26 rudybot: stamourv: Ok, thanks. The second best would be to get good expansion in the *Python* shell. For some reason, ipython does not do auto-complete very well. I've tried a little of bpython and it has some better autocompletion but I do not know if one can make it work from emacs
11:26 stamourv: anonymous97890: If you want rudybot to process multiple expressions, you need to explicitly ask it to `eval'.
11:27 anonymous97890: stamourv: thanks
11:27 stamourv: Otherwise, it thinks you're just talking to it, and tries to respond with something that's on topic.
11:27 stamourv: For some values of "on topic".
11:27 stamourv: rudybot: Which is why we love you.
11:27 rudybot: stamourv: The culture of people who love, care about and invent technology is partly characterized by a diligent skepticism of technology
11:28 anonymous97890: eval (define a nan.0) (values (number? a) (positive? a))
11:28 anonymous97890: rudybot: eval (define a nan.0) (values (number? a) (positive? a))
11:28 rudybot: anonymous97890: your sandbox is ready
11:28 rudybot: anonymous97890: error: nan.0: undefined; <NEWLINE>  cannot reference an identifier before its definition <NEWLINE>   in module: 'program
11:28 anonymous97890: rudybot: eval (define a +nan.0) (values (number? a) (positive? a))
11:28 rudybot: anonymous97890: ; Value: #t
11:28 (quit) ambrosebs: Ping timeout: 264 seconds
11:28 rudybot: anonymous97890: ; Value#2: #f
11:28 anonymous97890: (positive? +nan.0)
11:28 anonymous97890: rudybot: (positive? +nan.0)
11:28 rudybot: anonymous97890: ; Value: #f
11:31 anonymous97890: rudybot: eval (struct a-b (c)) (struct a (b-c)) (a-b-c (a-b 1))
11:31 rudybot: anonymous97890: error: a-b-c: contract violation <NEWLINE>   expected: a? <NEWLINE>   given: #<a-b>
11:31 anonymous97890: hm...
11:31 anonymous97890: rudybot: eval (struct a-b (c)) (struct a (b-c)) (a-b-c (a 1))
11:31 rudybot: anonymous97890: ; Value: 1
11:32 (join) ambrosebs
11:43 (quit) rohni: Quit: Leaving.
11:47 (join) bitonic
11:55 (quit) ambrosebs: Remote host closed the connection
11:57 (join) RacketCommitBot
11:57 RacketCommitBot: [racket] plt pushed 3 new commits to master: http://git.io/sVU8Eg
11:57 RacketCommitBot: racket/master ef189b8 Matthias Felleisen: Racket rename
11:57 RacketCommitBot: racket/master c13d603 Matthias Felleisen: fixed .ss name in title
11:57 RacketCommitBot: racket/master ec0398f Matthias Felleisen: Racket rename
11:57 (part) RacketCommitBot
12:02 (quit) mizu_no_oto: Quit: ["Textual IRC Client: www.textualapp.com"]
12:05 (join) mithos28
12:05 (join) jonrafkind
12:18 (quit) didi: Remote host closed the connection
12:21 (quit) egnarts-ms: Ping timeout: 245 seconds
12:21 (join) RacketCommitBot
12:21 RacketCommitBot: [racket] plt pushed 3 new commits to master: http://git.io/8s4Zdw
12:21 RacketCommitBot: racket/master 39ae7a8 Jay McCarthy: Removing saltification compat
12:21 RacketCommitBot: racket/master 9d3a42f Jay McCarthy: using bcrypt
12:21 RacketCommitBot: racket/master 6bf03c1 Jay McCarthy: Extending cookie support with expiration...
12:21 (part) RacketCommitBot
12:27 (join) RacketCommitBot
12:27 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/YRqb_A
12:27 RacketCommitBot: racket/master bb17b6a Asumu Takikawa: Add take-while, drop-while to racket/list
12:27 (part) RacketCommitBot
12:29 (join) dauterive
12:30 (join) anRch
12:30 (quit) codeblack: Ping timeout: 248 seconds
12:36 (join) mceier
12:37 (join) dsantiago
12:39 (join) Shozan
12:41 (quit) dented42: Ping timeout: 252 seconds
12:44 (join) mattmight
12:45 (join) soegaard_
12:45 (quit) soegaard_: Client Quit
12:45 (quit) soegaard: *.net *.split
12:45 (quit) Skola: *.net *.split
12:45 (quit) SHODAN: *.net *.split
12:45 (quit) paddymahoney: *.net *.split
12:45 (quit) mattmigh`: *.net *.split
12:45 (join) Skola
12:48 (join) ASau
12:48 (join) dented42
12:50 (join) didi
12:52 (join) paddymahoney
12:59 (join) mizu_no_oto
13:00 (quit) jbclements: Quit: jbclements
13:02 (quit) Fare: Read error: Operation timed out
13:02 (quit) anRch: Quit: anRch
13:03 (quit) didi: Remote host closed the connection
13:04 (join) jbclements
13:12 (join) didi
13:14 (join) soegaard
13:19 (join) ynniv
13:20 (quit) paddymahoney: Remote host closed the connection
13:32 (join) cnonim
13:32 (quit) jbclements: Quit: jbclements
13:33 cnonim: hello, all, i have two function https://gist.github.com/cNoNim/1877036b61637c6a18b6 ...
13:33 cnonim: how i can rewrite the get-char using get-any?
13:34 (quit) mithos28: Quit: mithos28
13:36 (part) ynniv
13:36 stamourv: cnonim: You'd either need to have some `peek' functionality, that allows you to look at the input without consuming it, or to have a way to pass a predicate to `get-any'.
13:38 cnonim: predicate it's next? or not?
13:39 (quit) jeapostrophe: Ping timeout: 252 seconds
13:40 stamourv: Since your example doesn't show what `next' can be, I don't know.
13:45 (join) rich2
13:46 cnonim: I think that the right decision here https://gist.github.com/cNoNim/1877036b61637c6a18b6
13:47 cnonim: but I'm not sure, and that can be simplified?
13:50 cnonim: i found best solution :)
13:51 cnonim: https://gist.github.com/cNoNim/1877036b61637c6a18b6
13:53 rich2: is there something like a range type in racket - which guarantees an instance of that type is between a lower and upper bound? or should that be a custom struct?
13:54 (quit) soegaard: Quit: soegaard
13:57 (join) anRch
13:58 stamourv: rich2: A custom struct could work, or you could use contracts.
14:00 eli: asumu: ping
14:01 (join) dyoo
14:10 (join) jeapostrophe
14:11 (quit) bitonic: Read error: Operation timed out
14:13 (quit) anonymous97890: Ping timeout: 264 seconds
14:19 (quit) yacks: Read error: Connection reset by peer
14:22 (join) jbclements
14:27 (join) rohni
14:43 asumu: eli: pong
14:44 asumu: rich2: I would just use (between/c low hi) to enforce this.
14:44 eli: asumu: As long as your in that part of my todo list, how about the rest of the neighborhood?
14:44 asumu: eli: are you referring to srfi-1?
14:45 eli: No.
14:45 eli: I have some things on a todo list that I never get to.
14:45 eli: You've done two, and there's a bunch of similar ones.
14:45 eli: `take-right-while', `drop-right-while'
14:46 eli: Also `split-at-while', `split-at-right-while'
14:46 asumu: I could add those. You think span from srfi-1 is worth it?
14:46 eli: What was it doing?
14:46 asumu: span = (values (take-while pred lst) (drop-while pred lst))
14:47 eli: Ah, that horrible name...
14:47 eli: That would be `split-at-while'
14:47 asumu: break = (values (take-while (negate pred) lst) (drop-while (negate pred) lst))
14:47 asumu: Oh, right.
14:47 asumu: Yeah, split-at is a better name.
14:47 eli: rudybot: (take 4 '(1 2 3 4 5 6 7 8))
14:47 rudybot: eli: your sandbox is ready
14:47 rudybot: eli: error: take: contract violation <NEWLINE>   expected: exact-nonnegative-integer? <NEWLINE>   given: '(1 2 3 4 5 6 7 8) <NEWLINE>   argument position: 2nd <NEWLINE>   other arguments...: <NEWLINE>    4
14:47 asumu: (if somewhat wordy in this case)
14:47 eli sighs
14:47 eli: rudybot: (take '(1 2 3 4 5 6 7 8) 4)
14:47 rudybot: eli: ; Value: (1 2 3 4)
14:47 eli: rudybot: (split-at '(1 2 3 4 5 6 7 8) 4)
14:47 rudybot: eli: ; Value: (1 2 3 4)
14:47 rudybot: eli: ; Value#2: (5 6 7 8)
14:48 eli: So it's the same.
14:48 eli: rudybot: (take-right '(1 2 3 4 5 6 7 8) 2)
14:48 rudybot: eli: ; Value: (7 8)
14:48 eli: rudybot: (split-at-right '(1 2 3 4 5 6 7 8) 2)
14:48 rudybot: eli: ; Value: (1 2 3 4 5 6)
14:48 rudybot: eli: ; Value#2: (7 8)
14:49 eli: I think that the reason that `span' and `break' were used is because they fit some random use case that happened to be relevant at the time.
14:49 eli: asumu: But I have a style commit in a few minutes.
14:52 asumu: How about splitf for split-at-while?
14:53 (join) RacketCommitBot
14:53 RacketCommitBot: [racket] plt pushed 4 new commits to master: http://git.io/Ez_PQg
14:53 RacketCommitBot: racket/master 5c90a7b Robby Findler: adjust gc2's mutator language so that void-producing expressions...
14:53 RacketCommitBot: racket/master 49ffca8 Robby Findler: remove ->address
14:53 RacketCommitBot: racket/master 9ef439d Robby Findler: add in documentation for plai/gc2
14:53 (part) RacketCommitBot
14:54 eli: asumu: The problem is that there's already an established naming -- it's good to have a direct connection between `split-at' and `split-at-while'.
14:55 (join) mithos28
14:58 (quit) cored: Read error: Operation timed out
14:59 (quit) mizu_no_oto: Quit: ["Textual IRC Client: www.textualapp.com"]
15:00 (join) soegaard
15:01 stamourv: eli: `split-at-while' is a terrible name. It implies that you're splitting multiple times, which it doesn't do.
15:02 (join) cored
15:02 (quit) cored: Changing host
15:02 (join) cored
15:02 eli: `splitf' is terrible for a more important reason, IMO.
15:03 eli: And if you're talking about *that* problematic interpretation of "while", then `take-while' is suffering from the same confusion -- and that's easy to change now before more code grows.
15:03 stamourv: IMO, `at' implies that you're passing it an index, which is also not the case.
15:03 eli: For example, `takef' -- which would make `splitf' better.
15:04 stamourv: Re `take-while': That's not as bad, since you're taking elements (in the english sense) repeatedly.
15:05 eli: I care much less about the english than I care about the identifiers, which is why I will fight for related names.
15:06 stamourv: How useful is taht function anyway?
15:06 (quit) anRch: Quit: anRch
15:06 asumu: Heh, turns out Haskell uses `span`, `break`. Clojure `split-with`.
15:07 stamourv: Does `split-with' do the same thing?
15:08 asumu: As `split-at-while`/`span`? Yes.
15:08 asumu: http://clojuredocs.org/clojure_core/clojure.core/split-with
15:09 eli: asumu: Yes, I have this comment in my todo file: bad name, but it's symmetric; clojure calls this `split-with'
15:09 eli: (So it's not like I didn't think about it before.)
15:09 stamourv: `split-with' sounds like partitioning to me, not scanning the list as long as a condition holds.
15:09 asumu: Yeah, I don't think `span` or `split-with` are worth considering.
15:10 stamourv: eli: Unrelated, are you in? I want to bounce an idea off you.
15:10 (quit) mithos28: Quit: mithos28
15:12 asumu: eli: what is the semantics of `take-right-while`? Is it just (reverse (take-while pred (reverse lst)))?
15:13 eli: asumu: Yes, but I hope that you won't implement it this way...
15:13 asumu: I wasn't planning to.
15:13 eli: The implementation will be a tiny bit tricky, with two pointers, one at the current position and one at the beginning of the chain.
15:13 eli: stamourv: No, I'm not in.
15:14 eli: stamourv: Re the name -- that's the problem I had with `split-with', which is why I didn't use it.
15:14 stamourv: eli: Hmm, maybe we can do it over IRC, then. Would you prefer to wait until the current conversation is over?
15:14 eli: So overall, you can look for some better word than `while' that might fir both.
15:14 eli: s/fir/fit/
15:14 stamourv: eli: Re `split-with': I agree.
15:15 eli: And again -- it has the advantage of `-while' not being a proper convention -- yet.
15:15 eli: so maybe a *f or -pred or something like that.
15:15 eli: And BTW², just to show how much I care about this: the problem of names is the only reason I didn't add these things yet.
15:15 stamourv: I like *f, it's consistent with `memf', `assf', etc.
15:16 eli: IIRC, I've asked the list a few times.
15:16 stamourv: Well, these functions aren't all that used either, which may explain the lack of interest.
15:16 eli: I think that people wanted the -while variants a few times.
15:16 stamourv: IMO, racket/list doesn't absolutely need feature parity with srfi/1.
15:16 eli: And the point of `break' and `span' definitely came up.
15:17 eli: (With the obvious clashes with html-like code and loops.)
15:17 stamourv: It just needs enough that we don't ever need to use srfi/1 and its horrible names ever again.
15:17 stamourv pukes in `iota''s general direction.
15:17 (join) anonymous97890
15:17 asumu: So, while I don't have strong feelings on whatever `span`/`split-at-while` is, I think we should use `take-while`/`drop-while`.
15:17 asumu: Because these are more or less universal
15:17 eli: Yeat, iota is a horrible name
15:17 asumu: (see OCaml, Haskell, Clojure)
15:18 eli: And in racket it's a little more horrible, since it should have been ι (but that's still bad).
15:18 stamourv: `take-while' and `drop-while' are fine with me, but I agree with eli that having a convention that would fit with the others would be nice.
15:18 eli: stamourv: Email is probably better for the other thing.
15:18 stamourv: Which is why I like *f.
15:18 stamourv: eli: Re other thing: Sounds good.
15:18 eli: Yeah, I think that the *f thing would work nicer.
15:18 (join) bitonic
15:19 eli: asumu: Maybe just dump a list of new names on dev and see if there are objections?
15:20 asumu: I can do that, but what is the current favorite? splitf? split-pred?
15:20 eli: I think that both me and stamourv lean towards the *f thing
15:21 stamourv: eli: Re other thing: Actually, I think waiting to talk in person would be better. Are you going to be in tomorrow?
15:21 eli: ("splitf" might be dangerous when talking about it.)
15:21 eli: (Especially during dinners.)
15:21 soegaard: Does this have another name? https://github.com/soegaard/this-and-that/blob/master/split-between.rkt
15:21 eli: stamourv: Maybe call?
15:21 stamourv: eli: Nah, not urgent.
15:22 asumu: So take-while, drop-while, take-while-right, drop-while-right, splitf, splitf-right.
15:23 eli: asumu: no, takef, dropf, takef-right, dropf-right, splitf, splitf-right
15:23 stamourv: (add1 eli)
15:23 eli: (The first "no" is not a name...)
15:23 eli: soegaard: Other things I have on my todo list are kind of similar, but not exactly
15:23 (quit) vkz: Quit: vkz
15:23 asumu: Okay.
15:23 eli: soegaard: I have: Something like clojure's `partition-by' and `group-by'
15:24 asumu: (while I like take-while better than takef, takef *is* nice and short...)
15:24 soegaard: I'll look them up
15:24 eli: soegaard: I also have `split-by'
15:24 stamourv: eli, soegaard: unstable/list has `group-by'. I use it fairly often.
15:24 eli: WHich is this: (split-by '(1 2 3 x 4 5 6 x 7) 'x) => '((1 2 3) (4 5 6) (7))
15:25 eli: stamourv: That group-by is also one of them, which I think has a more conventional name.
15:25 eli: (I think.  Don't have time to look it up now.)
15:25 stamourv: eli: I took the name from SQL, which IMO is a pretty authoritative source.
15:25 eli: soegaard: Also, you refered yesterday to some useful things still in srfi-13?
15:26 eli: stamourv: I think that generally speaking clojure should be more important for us.
15:26 eli: (Ie, there were several complaints in the spirit of "clojure has X, how do I do it in Racket" -- but that never happened with sql.)
15:26 soegaard: eli: string-prefix? and string-suffix? and string-contains?
15:27 stamourv: eli: Agreed, but IMO sharing the name with SQL makes the function easily discoverable.
15:27 eli: soegaard: I have `string-index' on my list, which covers the last and the first, not sure what to do about the -suffix one though...
15:28 soegaard: ok
15:28 eli: stamourv: Wait, your `group-by' is different than the one in clojure, right?
15:29 eli: (Again, apologies for not looking things up.)
15:29 stamourv: eli: I don't know what Clojure's `group-by' does.
15:29 stamourv: Mine groups things by equivalence class.
15:29 eli: stamourv: http://clojuredocs.org/clojure_core/clojure.core/group-by
15:29 stamourv: rudybot: (require unstable/list)
15:29 rudybot: stamourv: Done.
15:29 stamourv: rudybot: (group-by (lambda (x y) (= (mod x 3) (mod y 3))) (range 12))
15:29 rudybot: stamourv: error: mod: undefined; <NEWLINE>  cannot reference an identifier before its definition <NEWLINE>   in module: 'program
15:30 stamourv: rudybot: (group-by (lambda (x y) (= (modulo x 3) (modulo y 3))) (range 12))
15:30 rudybot: stamourv: ; Value: ((10 7 4 1) (11 8 5 2) (9 6 3 0))
15:30 eli: stamourv: Can you just look that up?
15:30 stamourv: Looking now.
15:31 asumu: rudybot: eval (require unstable/list) (group-by string-length '("a" "as" "asd" "aa" "asdf" "qwer"))
15:31 rudybot: asumu: your sandbox is ready
15:31 rudybot: asumu: error: string-length: arity mismatch; <NEWLINE>  the expected number of arguments does not match the given number <NEWLINE>   expected: 1 <NEWLINE>   given: 2 <NEWLINE>   arguments...: <NEWLINE>    "as" <NEWLINE>    "a"
15:31 asumu: It's not quite the same.
15:31 stamourv: Yeah, not exactly.
15:31 stamourv: rudybot: eval (require unstable/list) (group-by = #:key string-length '("a" "as" "asd" "aa" "asdf" "qwer"))
15:31 rudybot: stamourv: ; Value: (("qwer" "asdf") ("asd") ("aa" "as") ("a"))
15:31 stamourv: IMO more flexible, though.
15:32 stamourv: Output format is not exactly the same, but I like mine better.
15:32 asumu: stamourv: the Clojure version also gives you the output of the key function.
15:33 stamourv: I know, but I don't think it's a good idea.
15:33 asumu: Yeah, I'm not saying it's a good idea. Just another difference.
15:33 stamourv: You can end up with different keys that happen to be equivalent, given your equivalence predicate, so you'd need to elect a representative, etc.
15:33 eli: stamourv: So wait, your code applies N² predicates, or applies them to "some pairs to determine classes"?
15:33 eli: Either of these would be a bad idea.
15:34 eli: I have a vague memory of the latter.
15:34 stamourv: I forget the implementation. Let me look it up.
15:36 stamourv: O(N-elts * N-classes)
15:37 stamourv: So O(N²) in the worst case.
15:38 eli: stamourv: So, I was right -- the latter.
15:38 eli: That makes it a *very* odd addition.
15:38 stamourv: Why?
15:38 stamourv: IMO, it serves a useful purpose. I use it often.
15:40 eli: It requires an equivalence relation, but it's practically never useful with the built in obvious ones.  In most cases it would be some obvious composition of (lambda (x y) (some-equal? (some-key x) (some-key y))), which would make the clojure version sufficient.
15:41 eli: In any case, I think that clojure should definitely win here -- it's definitely much more paid-attention-to.
15:42 stamourv: eli: So you'd have both the key and the equivalence relation be optional/keywords?
15:42 stamourv: That sounds reasonable.
15:43 (join) Kaylin
15:46 eli: stamourv: Well, in the clojure version the key function is the main argument, and we can add a comparison keyword thing.  (Or maybe a last argument.)
15:51 stamourv: IME, the equivalence arguments was useful more often than the key argument.
15:58 rich2: asumu: stamourv: thanks. didn't know about between/c
15:58 rich2: although a general between seems useful and more readable than chaining >= and <=
15:59 stamourv: rudybot: (<= 2 3 4)
15:59 rudybot: stamourv: ; Value: #t
15:59 rich2: (between 5 1 10) => #t, (between 55 1 10) => #f
15:59 stamourv: rudybot: (define (between x y z) (<= y x z))
15:59 rudybot: stamourv: Done.
15:59 stamourv: rudybot: (between 5 1 10)
15:59 rudybot: stamourv: ; Value: #t
15:59 stamourv: rudybot: (between 55 1 10)
15:59 rudybot: stamourv: ; Value: #f
15:59 stamourv: rich2: ^
16:00 rich2: cool, i didn't realise that about <=
16:00 stamourv: A lot of Racket functions take an arbitrary number of arguments.
16:01 stamourv: rudybot: (+ 1 2 3 4)
16:01 rudybot: stamourv: ; Value: 10
16:01 stamourv: rudybot: (map list (range 5) (range 6 10))
16:01 rudybot: stamourv: error: map: all lists must have same size; arguments were: #<procedure:list> '(0 1 2 3 4) '(6 7 8 9)
16:01 stamourv: rudybot: (map list (range 5) (range 6 11))
16:01 rudybot: stamourv: ; Value: ((0 6) (1 7) (2 8) (3 9) (4 10))
16:02 stamourv: So a lot of things that, in other languages would need to be separate functions (e.g. `map' and `zipWith' in Haskell) are a single function in Racket.
16:03 rich2: how would haskell do (map list lst1 lst2) ?
16:04 rich2: that's a bit like the for/list having 2 clauses
16:04 stamourv: Something like `zipWith (\ x y . [x y]) lst1 lst2'.
16:04 stamourv 's Haskell is rusty.
16:05 stamourv: Re `for/list': Exactly!
16:05 stamourv: TBH, unless I'm mapping a function that already exists, I tend to use `for/list'.
16:07 (join) Mordae
16:07 Mordae: Hello.
16:07 (quit) anonymous97890: Ping timeout: 240 seconds
16:09 asumu: Mordae: hi
16:09 Mordae: I am playing with Racket's FFI with a short-term goal of wrapping libgirepository-1.0.so or so.
16:10 Mordae: There is one thing that puzzles me. Racket have support for C-style-memory-layout backed structures via (define-cstruct) and others.
16:10 Mordae: This mechanism directly supports inheritance, which is something really cool.
16:11 Mordae: But I have no idea how to wrap opaque structures that need only the type-compatibility part of inheritance.
16:14 Mordae: Is everything going to be fine if I just (cpointer-push-tag!) for every parent in the chain? How effective is that?
16:15 cnonim: how i can bind some argument to function?
16:17 dyoo: cnonim: do you have an example of what you're trying to do?
16:17 asumu: Mordae: I'm not quite sure what you mean by type-compatibility part. You mean you want something that acts like _tag but doesn't actually include _tag as a field? (like define-cstruct does)
16:19 cnonim: dyoo: have (lambda (x) x) and 1, do (foo (lambda (x) x) 1) and get (lambda () ((lambda (x) x) 1))... what's foo?
16:20 (quit) sizz_: Ping timeout: 245 seconds
16:20 dyoo: that looks like curry, right?
16:21 dyoo: http://docs.racket-lang.org/reference/procedures.html#%28def._%28%28lib._racket%2Ffunction..rkt%29._curry%29%29
16:21 rudybot: http://tinyurl.com/c9e3zxf
16:22 stamourv: rudybot: (define (foo f x) (lambda () (f x)))
16:22 rudybot: stamourv: Done.
16:22 stamourv: rudybot: (foo (lambda (x) x) 1)
16:22 rudybot: stamourv: ; Value: #<procedure>
16:22 stamourv: rudybot: ((foo (lambda (x) x) 1))
16:22 rudybot: stamourv: ; Value: 1
16:22 stamourv: cnonim: ^
16:22 stamourv: Functions are values, you can pass them as arugments, just like numbers.
16:23 Mordae: asumu: I am not sure I follow. As I understand tags, they define pointer type compatibility. A pointer tagged with 'foo and 'bar can be passed to a function accepting 'foo but not to a function accepting 'apples. (define-cstruct) seems to be doing something really cool behind the scenes making child structs being compatible with parents (appear tagged as such?). This is something I want, but can't have since the structs I deal with are
16:23 Mordae: opaque.
16:24 cnonim: i have list of arguments and (lambda (x) x) how i can get list of (lambda () ...)?
16:24 asumu: Mordae: it's not doing anything too complicated. If you have a parent type _type and want to define a new pointer type that can be used as a _type, you can do (define-cpointer-type _new-type _type).
16:24 dyoo: You can "thunk" a function.
16:24 stamourv: cnonim: (map foo list-of-lambdas list-of-args)
16:25 cnonim: stamourv: not list of lambda only one labda
16:25 cnonim: *lambda
16:26 rich2: (map your-lambda list-of-args) ?
16:26 dyoo: rudybot: (define (thunk a-val) (lambda () a-val))
16:26 rudybot: dyoo: your sandbox is ready
16:26 rudybot: dyoo: Done.
16:26 dyoo: rudybot: (define x-thunk (thunk 5))
16:26 rudybot: dyoo: Done.
16:26 dyoo: rudybot: x-thunk
16:26 rudybot: dyoo: ; Value: #<procedure>
16:26 stamourv: (map (lambda (x) (foo your-lambda x)) list-of-xs)
16:26 dyoo: rudybot: (x-thunk)
16:26 rudybot: dyoo: ; Value: 5
16:27 cnonim: stamourv: thx, it's logical :)
16:27 (quit) dyoo: Quit: dyoo
16:28 stamourv: Racket, it just makes sense.™
16:28 cnonim: rudybot: (map (lambda (x) (curry (lambda (y) y) x)) '(1 2 3 4 5)
16:28 Mordae: asumu: I totally missed that part of the docs. Thanks a lot!
16:28 rudybot: cnonim: So... Curry curried schönfinkeling?
16:28 cnonim: rudybot: (map (lambda (x) (curry (lambda (y) y) x)) '(1 2 3 4 5))
16:28 rudybot: cnonim: your sandbox is ready
16:28 rudybot: cnonim: ; Value: (1 2 3 4 5)
16:28 stamourv: cnonim: Missing a paren.
16:28 cnonim: yes, yes
16:29 stamourv: Also, you don't need `curry' there.
16:29 cnonim: why resul of function is values? not function?
16:29 stamourv: rudybot: (map (lambda (x) ((lambda (y) y) x)) '(1 2 3 4 5))
16:29 rudybot: stamourv: ; Value: (1 2 3 4 5)
16:30 stamourv: rudybot: (map (lambda (x) (lambda () ((lambda (y) y) x))) '(1 2 3 4 5))
16:30 rudybot: stamourv: ; Value: (#<procedure> #<procedure> #<procedure> #<procedure> #<procedure>)
16:30 stamourv: cnonim: ^
16:30 cnonim: thx )
16:32 (join) mithos28
16:34 cnonim: have (lambda (.x) ...) and list of args how pass args to lambda?
16:35 jonrafkind: ((lambda x ...) 1 2 3 4)
16:36 cnonim: not, we have '(1 2 3 4) and variadic lambda (lambda (.x) ...)
16:37 cnonim: *(lambda x ...)
16:39 cnonim: rudybot: ((lambda x (display x) 1 2 3 4)
16:39 rudybot: cnonim: lovely people please help... take a look at this.. http://paste.lisp.org/display/94865  ... how come emacs what the unicode character is but desides to display a completely different one?....
16:39 jonrafkind: (lambda x ...) is a function that accepts a list of arguments for x
16:40 cnonim: rudybot: ((lambda x (display x)) 1 2 3 4)
16:40 rudybot: cnonim: ; stdout: "(1 2 3 4)"
16:40 cnonim: rudybot: ((lambda x (display x)) '(1 2 3 4))
16:40 rudybot: cnonim: ; stdout: "((1 2 3 4))"
16:40 jonrafkind: rudybot: (apply (lambda x (display x)) '(1 2 3 4))
16:40 rudybot: jonrafkind: your sandbox is ready
16:40 rudybot: jonrafkind: ; stdout: "(1 2 3 4)"
16:40 cnonim: jonrafkind: thx
16:43 (join) RacketCommitBot
16:43 RacketCommitBot: [racket] plt pushed 3 new commits to master: http://git.io/ttaiMQ
16:43 RacketCommitBot: racket/master 95abacd Eli Barzilay: GCalc updates....
16:43 RacketCommitBot: racket/master bbeb293 Eli Barzilay: Fix renamed file in props.
16:43 RacketCommitBot: racket/master b9406e6 Eli Barzilay: Style changes....
16:43 (part) RacketCommitBot
16:43 eli: asumu: that was the change that I was talking about.
16:43 eli leaves
16:45 cnonim: https://gist.github.com/cNoNim/1877036b61637c6a18b6
16:45 cnonim: it works, thx guys :)
16:48 (quit) mithos28: Quit: mithos28
16:49 (quit) jonrafkind: Ping timeout: 276 seconds
16:51 (quit) vipjun: Ping timeout: 256 seconds
16:56 (quit) Skola: 
16:59 (quit) basdirks: Ping timeout: 245 seconds
17:03 (join) RacketCommitBot
17:03 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/kooLqA
17:03 RacketCommitBot: racket/master 308a65e Robby Findler: fix plai tests
17:03 (part) RacketCommitBot
17:06 (join) kvda
17:12 (quit) cnonim: Quit: WeeChat 0.4.0
17:12 (join) mizu_no_oto
17:12 didi: rudybot: People of Earth.
17:12 rudybot: didi: From humble beginnings on Mars, Bloom Energy is now changing the Earth for the better.
17:14 (quit) bjz: Quit: Leaving...
17:21 rich2: it seems the date struct can handle dates outside 32 bit unix time - but the other constructors such as seconds->date can't
17:22 (part) Mordae
17:25 rich2: i'll have to test if reading dates from an RDBMS will work which are before 1970
17:26 (quit) Kaylin: Read error: Connection reset by peer
17:27 (join) RacketCommitBot
17:27 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/Xs-l9Q
17:27 RacketCommitBot: racket/master 0720ad2 Jay McCarthy: raco setup, ignore this collect
17:27 (part) RacketCommitBot
17:28 (quit) serhart: Quit: Leaving.
17:28 (join) jonrafkind
17:28 (quit) rich2: Quit: Leaving.
17:34 (join) bjz
17:36 (quit) kvda: Quit: x___x
18:11 (join) ambrosebs
18:11 (join) paddymahoney
18:11 (join) zyoung_
18:13 (join) mithos28
18:13 (quit) zyoung_: Remote host closed the connection
18:14 (quit) zyoung: Ping timeout: 260 seconds
18:28 (quit) mithos28: Quit: mithos28
18:32 (quit) mizu_no_oto: Quit: Computer has gone to sleep.
18:38 (quit) sstrickl: Quit: sstrickl
18:47 (quit) soegaard: Quit: soegaard
18:59 (join) mizu_no_oto
19:11 (join) sstrickl
19:11 (quit) sstrickl: Changing host
19:11 (join) sstrickl
19:11 (quit) bjz: Quit: Leaving...
19:29 (quit) jonrafkind: *.net *.split
19:29 (quit) mceier: *.net *.split
19:29 (quit) arrdem: *.net *.split
19:30 (quit) mizu_no_oto: Quit: Computer has gone to sleep.
19:30 (quit) TheBeerinator: Ping timeout: 248 seconds
19:30 (join) chandler
19:30 (quit) chandler: Changing host
19:30 (join) chandler
19:31 (join) TheGhostinaor
19:36 (quit) sstrickl: Quit: sstrickl
19:39 (join) jonrafkind
19:39 (join) mceier
19:39 (join) arrdem
19:50 (join) mithos28
19:56 (quit) cored: Ping timeout: 272 seconds
19:58 (join) cored
19:58 (quit) cored: Changing host
19:58 (join) cored
20:02 (quit) jbclements: Quit: jbclements
20:05 (quit) mithos28: Quit: mithos28
20:12 (join) Kaylin
20:16 (join) mizu_no_oto
20:39 (quit) mizu_no_oto: Quit: Computer has gone to sleep.
20:42 (quit) jonrafkind: Ping timeout: 257 seconds
20:53 (quit) bitonic: Ping timeout: 260 seconds
20:55 (quit) jeapostrophe: Ping timeout: 240 seconds
21:01 (join) mizu_no_oto
21:03 (join) bitonic
21:13 (quit) bitonic: Ping timeout: 255 seconds
21:28 (join) mithos28
21:35 (join) DanBurton
21:43 (quit) mithos28: Quit: mithos28
21:44 (quit) mizu_no_oto: Quit: Computer has gone to sleep.
21:57 (quit) Kaylin: Read error: Connection reset by peer
22:16 (join) mizu_no_oto
22:20 (join) yacks
22:24 (join) Kaylin
22:29 (quit) didi: Quit: ERC Version 5.3 (IRC client for Emacs)
22:33 (quit) dauterive: Quit: Leaving
22:39 (join) zacts
22:40 (quit) zacts: Changing host
22:40 (join) zacts
22:41 (quit) DanBurton: Remote host closed the connection
22:46 (join) sizz
22:54 (join) zyoung
23:04 (quit) Kaylin: Quit: Leaving.
23:07 (join) mithos28
23:13 (quit) mizu_no_oto: Quit: Computer has gone to sleep.
23:22 (quit) mithos28: Quit: mithos28
23:23 (join) Fare
23:31 (join) noam_
23:34 (quit) noam: Ping timeout: 264 seconds
23:52 (join) jonrafkind
23:54 (join) Kaylin
23:54 (quit) karswell: Ping timeout: 264 seconds