00:01 (join) cdidd
00:14 (quit) jao: Ping timeout: 264 seconds
00:17 (join) kvda
00:24 (quit) brum: Remote host closed the connection
00:24 (join) rmathews
00:25 (join) ASau
00:26 (quit) dsantiago: Ping timeout: 260 seconds
00:29 (join) dsantiago
00:32 (join) Kaylin
00:35 (join) ambrosebs
00:35 (quit) vipjun: 
00:36 (quit) cdidd: Read error: Connection reset by peer
00:41 (join) cdidd
00:56 eli: rudybot: later tell someone something
00:56 rudybot: eli: Thank bpalmer: I'll work with it, will come back later if more questions
00:57 eli: cky: I don't think that offby1 implemented the later tell thing.
00:57 eli: cky: As for that minesweeper code, it's *ancient*...
00:59 eli: stamourv: That code-ayatollah thing is something I use for checking students (in the early homework), but I posted it hoping that someone would extend it and make it more generally applicable.  (There's no reason for it to be better for students, other than the particular set of checks that I made it do.)
01:00 eli: bremner: Did you do some quick racket+etc tutorial in your class?
01:00 eli: asumu: ping
01:11 (quit) sirdancealot: Quit: ragequit
01:12 (join) cantcode
01:12 eli: So everybody's dead?
01:16 (part) eli: "*sigh*"
01:16 (join) eli
01:16 eli sighs
01:19 (join) asvil
01:22 (join) brum
01:27 (quit) brum: Ping timeout: 272 seconds
01:31 jonrafkind heaves
01:31 (quit) yacks: Remote host closed the connection
01:32 (quit) mceier: Quit: leaving
01:33 (join) yacks
01:40 (join) basdirks
01:50 (join) RacketCommitBot
01:50 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/1R_lSg
01:50 RacketCommitBot: racket/master b84a7bc Eli Barzilay: Fix typo in `string-replace' keyword name.
01:50 (part) RacketCommitBot
02:00 (quit) rmathews: Quit: ...
02:06 (quit) basdirks: Ping timeout: 255 seconds
02:23 (join) brum
02:23 (join) jao
02:24 (quit) jao: Changing host
02:24 (join) jao
02:28 (quit) brum: Ping timeout: 272 seconds
02:47 (join) rohni
02:53 (quit) crundar: Quit: Leaving
02:56 (quit) jonrafkind: Ping timeout: 276 seconds
02:57 (join) mceier
03:06 (quit) ivan`: Ping timeout: 248 seconds
03:08 (quit) Kaylin: Quit: Leaving.
03:08 (join) ivan`
03:13 (join) rmathews
03:13 (quit) ASau: Excess Flood
03:13 (join) soegaard
03:20 (quit) gridaphobe: Remote host closed the connection
03:31 (join) basdirks
03:33 (quit) ambrosebs: Remote host closed the connection
03:33 (quit) rmathews: Ping timeout: 272 seconds
03:34 (join) gridaphobe
03:37 (join) rmathews
03:41 (join) tilde`
03:46 (join) rmathews_
03:47 (quit) rmathews: Ping timeout: 276 seconds
03:47 (nick) rmathews_ -> rmathews
03:47 (quit) gridaphobe: Remote host closed the connection
03:49 (quit) tilde`: Quit: kthxbai
03:49 (quit) bjz: Quit: Bye!
03:49 (join) hkBst
03:50 (join) RacketCommitBot
03:50 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/4-Q5zg
03:50 RacketCommitBot: racket/master ca304aa Eli Barzilay: New Racket version 5.3.3.7.
03:50 (part) RacketCommitBot
03:52 (join) bjz
03:54 (quit) hkBst: Remote host closed the connection
03:54 (join) hkBst
03:54 (quit) hkBst: Changing host
03:54 (join) hkBst
04:15 (join) egnarts-ms
04:18 (join) gridaphobe
04:22 egnarts-ms: in Racket reference, it is noted that if "in-list", "in-range", etc. sequence functions appear directly in "for" iteration clause, this may increase performance
04:22 egnarts-ms: i guess this is because those identifiers have compile-time (phase 1) bindings ?
04:23 (quit) soegaard: Quit: soegaard
04:23 egnarts-ms: and syntax transformer of "for" can figure out something (at compile time)
04:24 mithos28: egnarts-ms: Sorta, they don't have phase1 bindings but phase1 looks at the phase0 bindings
04:27 (quit) gridaphobe: Ping timeout: 276 seconds
04:27 egnarts-ms: mithos28: this means that there's no way for Racket programmer to describe some sequence in phase 1 terms, right ?
04:28 mithos28: egnarts-ms: I'm not sure what you are asking.
04:29 mithos28: you could bring in racket/sequence at phase1 and use them there, but that is totally seperate from using it at phase 0
04:29 egnarts-ms: mithos28: i mean, by telling at compile time how to iterate vs. figuring out everything at run-time (AFAIU, as by using make-do-sequence, which is a run-time generator of sequences)
04:30 egnarts-ms: mithos28: for example, i wanna iterate through naturals like this: 1, n, 2, n-1, 3, n-2, etc.
04:30 (quit) hkBst: Ping timeout: 246 seconds
04:31 mithos28: you can use define-sequence-syntax
04:31 mithos28: which is how the built in ones are implemented
04:31 (join) caterpillarsandw
04:32 (quit) kvda: Quit: z____z
04:34 (join) finishingmove
04:34 (part) finishingmove
04:34 egnarts-ms: mithos28: ah, thank you for pointing to this.. Racket contains really a lot of stuff :)
04:34 (join) hkBst
04:36 (quit) racycle: Quit: racycle
04:36 caterpillarsandw: I get this error when I try to (load) a file: "#%top-interaction: unbound identifier;". Any idea how to fix it?
04:37 mithos28: caterpillarsandw: Don't use load, and save yourself a ton of pain
04:37 caterpillarsandw: Why? There aren't really any other alternatives when writing r5rs-compliant code.
04:38 (join) bitonic
04:38 mithos28: to solve the current issue you need to bind #%top-interaction
04:38 mithos28: #lang r5rs?
04:39 caterpillarsandw: Yeah, I'm using that.
04:40 mithos28: and what is the issue?
04:40 mithos28: You are loading code from that file?
04:41 caterpillarsandw: Well, I have a file written in r5rs from a library I wrote, and I'm trying to load it with drracket. Only the code that's trying to load it has the "#lang r5rs" line.
04:42 mithos28: Did you read the warning about loading on the page that describes r5rs
04:42 mithos28: #lang r5rs*
04:42 mithos28: load works the way you would expect it to in racket not r5rs
04:43 caterpillarsandw: I haven't, but I'll check it out now.
04:43 mithos28: You probably don't want #lang r5rs
04:43 mithos28: you probably want to write module free code
04:44 mithos28: so just 'racket -l r5rs/init -f <your-file>'
04:45 mithos28: without a #lang r5rs
04:46 mithos28: I would not recommend writing such code, but if you are using an existing library it should work
04:49 caterpillarsandw: That seems to work. Thanks.
05:02 asvil: it seems that last racket commit fails on tests/racket/optimize.rktl
05:03 (quit) jao: Ping timeout: 272 seconds
05:21 (quit) neilv: Ping timeout: 264 seconds
05:24 (quit) caterpillarsandw: Quit: Lost terminal
05:25 (join) brum
05:30 (quit) brum: Ping timeout: 272 seconds
05:34 (quit) rohni: Quit: Leaving.
05:39 basdirks: In Haskell when we create a variation of a function we usually temporarily rename it by appending a single quote
05:39 basdirks: f, f'
05:39 basdirks: what's the convention in racket/scheme?
05:51 hkBst: basdirks: a * can be used, though it's a weak convention
05:51 hkBst: weak, because it also carries the connotation of repetition
05:51 hkBst: as in let let*
05:52 basdirks: yeh
05:52 hkBst: basdirks: most alphanumerics are available for use in identifiers though
06:07 (join) rohni
06:17 (join) zacts
06:19 (join) vkz
06:32 (quit) merijn: Ping timeout: 248 seconds
06:35 (quit) egnarts-ms: Ping timeout: 245 seconds
06:36 asvil: sorry, optimize.rktl fails only on racketcgc
06:38 (join) ambrosebs
06:51 (join) noelw
07:15 (join) jeapostrophe
07:26 (quit) yacks: Ping timeout: 264 seconds
07:26 bremner: asvil: as part of the normal build process?
07:27 asvil: bremner: no, tests do not run as part of build process
07:27 (quit) ambrosebs: Remote host closed the connection
07:27 bremner: oh, ok.
07:27 asvil: I use tests for keeping my racket changes valid
07:28 (join) ambrosebs
07:28 (join) basdirks_
07:29 (join) yacks
07:31 (join) mceier_
07:31 (join) m4burns_
07:33 (quit) mceier: Disconnected by services
07:33 (nick) mceier_ -> mceier
07:35 (quit) basdirks: *.net *.split
07:35 (quit) m4burns: *.net *.split
07:50 (quit) rohni: Quit: Leaving.
08:10 (join) brum
08:12 (join) RacketCommitBot
08:12 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/BXGjFQ
08:12 RacketCommitBot: racket/master 8a08cfc Robby Findler: adjust plai test suite tests to check the ports that things now go to
08:12 (part) RacketCommitBot
08:13 (join) mizu_no_oto
08:14 (join) didi`
08:15 (nick) didi` -> didi
08:15 (quit) noelw: Ping timeout: 255 seconds
08:19 (quit) hkBst: Remote host closed the connection
08:19 (join) hkBst
08:21 (join) jeapostr1phe
08:25 (quit) hkBst: Read error: Connection reset by peer
08:28 (join) hkBst_
08:28 (quit) hkBst_: Changing host
08:28 (join) hkBst_
08:30 (quit) jeapostr1phe: Ping timeout: 264 seconds
08:31 (quit) bitonic: Ping timeout: 240 seconds
08:36 (quit) brum: Remote host closed the connection
08:39 (join) rmathews
08:54 (join) RacketCommitBot
08:54 RacketCommitBot: [racket] plt pushed 3 new commits to master: http://git.io/q2LJrQ
08:54 RacketCommitBot: racket/master 7734966 Matthew Flatt: `equal?' hashing: shortcut and some caching for structure transparency
08:54 RacketCommitBot: racket/master 7a8c2ff Matthew Flatt: improve complexity of `hash-iterate-{key,value}'...
08:54 RacketCommitBot: racket/master 39aafe3 Matthew Flatt: change `eq?' hashing to avoid dropping the low bit for fixnums
08:54 (part) RacketCommitBot
08:55 (join) ambroseb_
08:59 (quit) ambrosebs: Ping timeout: 264 seconds
08:59 (join) vkz_
09:00 (quit) rmathews: Read error: Connection reset by peer
09:00 (quit) mizu_no_oto: Quit: Computer has gone to sleep.
09:01 (quit) vkz: Ping timeout: 272 seconds
09:01 (nick) vkz_ -> vkz
09:05 (join) carleastlund
09:05 (join) rmathews
09:11 (quit) ambroseb_: Remote host closed the connection
09:14 (join) brum
09:14 (join) mizu_no_oto
09:15 (quit) bjz: Read error: Connection reset by peer
09:19 (quit) brum: Ping timeout: 244 seconds
09:23 asumu: eli: pong
09:25 (join) bitonic
09:30 (quit) mizu_no_oto: Quit: Computer has gone to sleep.
09:35 eli: asumu: Your change was too simplistic.
09:36 eli: The problem is that you made the LHS column also specified with a negative margin, so on a narrow screen it would get out of view with no way of scrolling left to it.
09:37 eli: It took me some time but I managed to get around this by adding a container around the whole thing with the right overall size.  (And you can see that in newdocs.racket-lang.org now.)
09:39 eli: As for the ,rr vs ,r -- the former uses a table etc, so it's not going through the usual channels of things like `only-in', and I don't think that there's a way to make it work.
09:39 eli: And re the zazzle thing -- I don't know if it should be up on the page, maybe that's a point to raise on the list, or something.
09:44 asumu: eli: oh okay. The version on newdocs looks good, though it's kinda weird that the "version-specific listing" box is now just empty.
09:45 eli: Yeah, that's actually a bug.
09:45 eli: There's a div with a `hide_when_on_the_web' class around the text, but the whole thing is inside a <p>, so the line is still there.
09:46 eli: I'll get rid of it too, though it makes the whole hide_when_on_the_web thing redundant...
09:47 asumu: Also I wonder if having "Racket documentation" on the web version is redundant.
09:47 asumu: Given that you have a Racket banner and you're on the documentation tab.
09:50 eli: I think that that's fine.
09:50 eli: It's like "Resources for learning" or a button that says "download" on the respective pages.
09:51 asumu: I suppose so.
09:51 asumu: eli: also for the ,rr thing, I wasn't actually wanting to use `only-in` and such. I just wanted to be able to do ,rr "foo.rkt"
09:52 eli: That should work, no?
09:52 asumu: It doesn't seem to, but maybe I'm doing something wrong.
09:52 eli: If it works with ,en then it should work with ,rr too.
09:54 asumu: Oh, nevermind, it works on another machine.
09:54 asumu: Maybe the racket on this machine is out of date.
09:54 eli happily ignores it...
09:58 cky: eli: :-) (Re minesweeper, and yeah, I know rudybot doesn't really have later tell, but maybe soegaard is watching logs. I guess now. :-P)
09:58 cky: eli: BTW, what do you think of this? https://github.com/cky/racket/commit/3f2bcc442f8827ab14fbd8a34c9e20e010c05dda
09:58 rudybot: ??
09:58 cky: eli: If it looks good to you, I'll send a formal pull request (or however you guys take patches).
10:01 eli: cky: Looks fine -- you tried+tested it, I assume?
10:01 eli: IOW, it's pull-ready, right?
10:01 (quit) asvil: Ping timeout: 272 seconds
10:03 cky: Yes, I've tested the code in http://stackoverflow.com/a/15109732/13 with it, both using #lang racket and #lang mzscheme.
10:04 (join) vkz
10:04 cky: If there are more things you think would be good to test, I can do that too.
10:06 (join) RacketCommitBot
10:06 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/XBLKcQ
10:06 RacketCommitBot: racket/master 9e93ee2 Chris K. Jester-Young: Convert srfi/61/cond to use modern module style....
10:06 (part) RacketCommitBot
10:06 eli: cky: I think that's fine -- of course, if you want to contribute tests, then lots of srfis need more of them...
10:07 cky: I'll see what I can do, when I find the time. Meanwhile, thanks for the push. :-)
10:14 (join) brum
10:19 (quit) brum: Ping timeout: 264 seconds
10:21 (join) mizu_no_oto
10:26 (join) ambrosebs
10:36 (join) anRch
10:38 (join) gridaphobe
10:46 stamourv: mithos28: ping
10:50 (quit) Fare: Ping timeout: 246 seconds
10:55 (join) soegaard
10:59 (join) Kaylin
11:00 (quit) hkBst_: Quit: Konversation terminated!
11:01 cky: soegaard: You could use something like (define log2 (let ((invlog2 (/ (log 2)))) (lambda (x) (* (log x) invlog2))), but I don't think there's anything offering access to the x87 instruction for doing log2.
11:01 soegaard: Thanks
11:02 soegaard: I think I asked the wrong question.
11:02 soegaard: I was wondering why the "#lang r6rs"  log takes 2 arguments and the racket one only one.
11:03 soegaard: I hope the optimzer lifts (log 2) automatically. Hmm. How can one check?
11:07 cky: I don't know. It's not like Racket has C++11 constexpr. ;-)
11:08 (join) jbclements
11:09 cky: soegaard: (eval `(define (log2 x) (* (log x) ,(/ (log 2)))))
11:09 cky: :-P
11:15 (join) brum
11:15 stamourv: soegaard: I'm pretty sure the Racket compiler would do that constant folding.
11:15 stamourv: To check, you can use raco decompile.
11:16 soegaard: stamourv: It does. I was just reading: http://docs.racket-lang.org/guide/performance.html
11:16 stamourv: Sadly, Optimization Coach doesn't report constant folding. Yet. :)
11:17 soegaard: cky: One of my favorite macro examples is  define-constant.
11:17 soegaard: It is simplest macro I know that requires propgation of information in phase 1.
11:18 (quit) gridaphobe: Remote host closed the connection
11:18 (join) racycle
11:18 jbclements: rudybot: (for/sum ([i (in-range 1 1000)]) (/ 1 (sub1 (* 16 i i))))
11:18 rudybot: jbclements: ; Value: 17428995982264011561162753415045345557728673910110907923451497974323927849609273608691673439998115825521790947205849074016472121433155960641958756492178165309577634724654424336300174523215089989974869313697003321037968938556532468410502533008154135115350210472139996595249644360678674922107018271840252586347456237355167652389964149972213893317964149222669208840782301669917262459334421699320780738194998694
11:20 jbclements: rudybot: (exact->inexact (for/sum ([i (in-range 1 1000)]) (/ 1 (sub1 (* 16 i i)))))
11:20 rudybot: jbclements: ; Value: 0.10723838703955514
11:20 (quit) brum: Ping timeout: 264 seconds
11:21 jbclements: can any math-folks here point me toward an analytic solution to the sum of the infinite series whose first 1K terms rudybot just summed for me? That is, sumof 1/16i^2 -1…. A wikipedia reference would be just dandy.
11:24 (quit) ambrosebs: Remote host closed the connection
11:24 (quit) basdirks_: Ping timeout: 256 seconds
11:29 (join) neilv
11:32 soegaard: jbclements: Plouffe's  inverter is usally the place to go:          http://pi.lacim.uqam.ca/
11:32 (join) Fare
11:32 jbclements: soegaard: thanks!
11:32 soegaard: It seems there is something wrong today though...
11:38 soegaard: jbclements: Wolfram Alpha can also find closed forms:   http://www.wolframalpha.com/input/?i=0.10723838703955514
11:40 soegaard: (4-pi)/8   ?
11:41 soegaard: http://www.wolframalpha.com/input/?i=sum+1%2F%2816n%5E2-1%29+from+1+to+infinity
11:41 rudybot: http://tinyurl.com/c9cm9cf
11:41 jbclements: soegaard: thanks! Wolfram alpha is … amazing.
11:42 (quit) anRch: Quit: anRch
11:45 soegaard: I think your problem is solved by the Wilf-Zeilberger algorithm. If you are interested the A=B book is amazing. http://www.math.upenn.edu/~wilf/AeqB.html
11:45 (join) bitonic`
11:46 (quit) bitonic: Read error: Connection reset by peer
11:46 (join) basdirks
11:47 (quit) Fare: Ping timeout: 264 seconds
11:49 (join) ryan_
11:49 jbclements: soegaard: thanks, you've just *completely* distracted me from my work…. my fault entirely, I suppose. :)
11:49 soegaard: :-)
11:54 soegaard: The A=B book is a really good read for anyone interested in computer algebra.
11:57 (quit) basdirks: Ping timeout: 264 seconds
12:01 (join) cnonim
12:01 cnonim: how i cant print syntax object?
12:03 soegaard: cnonim: I use display in DrRacket. Then I click the little triangle to see the contents.
12:04 cnonim: ups, not noticed
12:07 (join) RacketCommitBot
12:07 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/K8c8eg
12:07 RacketCommitBot: racket/master 2236363 Matthew Flatt: fix bug in hashing shortcut for transparency
12:07 (part) RacketCommitBot
12:09 cnonim: drracket can select s-expr quickly?
12:10 bremner: cnonim: select how? like destructure? maybe look at match
12:10 soegaard: shift + alt + arraows
12:10 bremner: oh, sorry, I missed "dr" ;)
12:12 cnonim: thx, guys ;) it's cool
12:13 cnonim: may be drracket can change type of bracket? example from '()' to '[]'
12:15 cnonim: or remove parenthesis? example have [(...long...)] how i can remove ()
12:15 (join) brum
12:17 (join) RacketCommitBot
12:17 RacketCommitBot: [racket] plt pushed 1 new commit to master: https://github.com/plt/racket/commit/5790667054568994d61c366b394be7deba28ce9a
12:17 RacketCommitBot: racket/master 5790667 Asumu Takikawa: Remove duplicate non-terminals in `define-extended-language`...
12:17 (part) RacketCommitBot
12:17 soegaard: You can find the keybindings in the edit menu.
12:20 (quit) brum: Ping timeout: 248 seconds
12:21 (quit) mizu_no_oto: Quit: ["Textual IRC Client: www.textualapp.com"]
12:24 (join) mceier
12:24 soegaard: cnonim:  ctrl+c ctrl+e
12:25 (quit) Kaylin: Quit: Leaving.
12:27 (join) basdirks
12:30 (nick) m4burns_ -> m4burns
12:46 (join) ASau
12:54 (quit) cnonim: Ping timeout: 255 seconds
12:59 (quit) neilv: Quit: Leaving
12:59 (join) jonrafkind
13:04 (quit) jeapostrophe: Ping timeout: 272 seconds
13:05 (join) jeapostrophe
13:07 (quit) bitonic`: Remote host closed the connection
13:07 (join) bitonic`
13:08 (join) mizu_no_oto
13:08 (join) Fare
13:09 (nick) Fare -> Guest58449
13:15 (join) rmathews_
13:16 (quit) rmathews: Ping timeout: 264 seconds
13:16 (nick) rmathews_ -> rmathews
13:16 (join) brum
13:19 (quit) zacts: Quit: ERC Version 5.3 (IRC client for Emacs)
13:21 (quit) brum: Ping timeout: 248 seconds
13:21 (join) zacts
13:22 (quit) yacks: Read error: Connection reset by peer
13:35 (join) rmathews
13:39 (join) rmathews_
13:40 (quit) rmathews: Ping timeout: 264 seconds
13:40 (nick) rmathews_ -> rmathews
13:48 (quit) whoops: Quit: Farewell
13:51 (quit) soegaard: Quit: soegaard
13:55 (join) whoops
13:59 (join) Guest52818
13:59 (quit) Guest52818: Changing host
13:59 (join) Guest52818
13:59 (quit) jeapostrophe: Ping timeout: 244 seconds
13:59 (quit) vkz: Quit: vkz
14:07 (join) vkz
14:09 (join) cnonim
14:16 (join) brum
14:21 (quit) brum: Ping timeout: 246 seconds
14:22 (join) didi
14:31 (join) merijn
14:38 cnonim: how in drracket find only words?
14:39 cnonim: example match next and not match next-result
14:39 cnonim: or regexp search may be
14:42 (join) soegaard
14:43 (quit) vkz: Quit: vkz
14:43 (join) RacketCommitBot
14:43 RacketCommitBot: [racket] plt pushed 2 new commits to master: http://git.io/i9viHA
14:43 RacketCommitBot: racket/master 542c38e Matthew Flatt: equality: struct transparency and inspector shortcuts and caching...
14:43 RacketCommitBot: racket/master ca43204 Matthew Flatt: fix missing initialization of cached inspector reference...
14:43 (part) RacketCommitBot
14:54 (join) dauterive
14:57 (join) vkz
14:57 (quit) vkz: Client Quit
15:01 (join) jeapostrophe
15:01 (quit) jeapostrophe: Changing host
15:01 (join) jeapostrophe
15:08 (quit) mizu_no_oto: Quit: ["Textual IRC Client: www.textualapp.com"]
15:09 (join) RacketCommitBot
15:09 RacketCommitBot: [racket] plt pushed 3 new commits to master: http://git.io/m48WdA
15:09 RacketCommitBot: racket/master 43912be Asumu Takikawa: Some tweaks for Guide section 4
15:09 RacketCommitBot: racket/master 48d0677 Asumu Takikawa: Improve Guide section 5
15:09 RacketCommitBot: racket/master 2065c82 Asumu Takikawa: Guide chapter 6 tweaks
15:09 (part) RacketCommitBot
15:11 (join) RacketCommitBot
15:11 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/W7hHRw
15:11 RacketCommitBot: racket/master 5e9d1ce Eli Barzilay: Use the `not-on-the-web' style *around* the installation comment....
15:11 (part) RacketCommitBot
15:13 (join) fasta
15:14 (join) mizu_no_oto
15:17 (join) brum
15:21 (quit) brum: Ping timeout: 248 seconds
15:23 (quit) cdidd: Remote host closed the connection
15:23 (join) dyoo
15:26 (quit) jbclements: Quit: jbclements
15:38 (quit) bitonic`: Remote host closed the connection
15:39 (join) BW^-
15:39 BW^-: hi guys - in this file,
15:39 BW^-: http://planet.plt-scheme.org/package-source/dherman/json.plt/4/0/main.rkt
15:40 BW^-: can you Please explain to me exactly how |for/haseq| in the |for/hasheq| procedure works?
15:40 BW^-: for/hasheq!
15:40 (join) gridaphobe
15:41 BW^-: isolated here: http://pastebin.ca/2326800
15:41 BW^-: soo, the for/hasheq has three arguments, at first the something-like "loop iterator" code,
15:41 BW^-: ([(key value)
15:41 BW^-: ...
15:41 BW^-: (values (string->symbol key) value))))
15:41 BW^-: (lambda (port)
15:41 BW^-: (eq? (peek-char port) #\})))])
15:41 BW^-: that one i understand
15:42 BW^-: but, how does the second and third arg to for/hasheq fit together?
15:42 stamourv: BW^-: `for/hasheq' iterates over a number of sequences, and returns an eq? hash-table.
15:42 BW^-: when are they invoked and of what effect are they?
15:42 stamourv: Let me show you an example.
15:42 stamourv: rudybot: init racjet
15:42 rudybot: stamourv: error: racjet: standard-module-name-resolver: collection not found <NEWLINE>   collection: "racjet" <NEWLINE>   in collection directories: <NEWLINE>    /mnt/racket-5.3.2/collects <NEWLINE>   in: racjet
15:42 stamourv: rudybot: init racket
15:42 rudybot: stamourv: your sandbox is ready
15:42 BW^-: stamourv: i'm trying to port it to gambit
15:43 stamourv: rudybot: (for/hasheq ([i (in-range 3)] [x (in-list '(a b c))]) (values i x))
15:43 rudybot: stamourv: ; Value: #hasheq((1 . b) (0 . a) (2 . c))
15:43 BW^-: stamourv: i ported all the rest to gambit but this little piece remains.
15:43 (quit) jonrafkind: Ping timeout: 248 seconds
15:43 BW^-: stamourv: right, great, i get that
15:43 BW^-: stamourv: now what about for/haseq:s second and third args??
15:43 BW^-: i get completely lost on those two!
15:44 stamourv: There are not really arguments, `for/haseq' is not a function.
15:44 BW^-: yeah but a form or macro right
15:44 stamourv: Anything beyond the first subform (the sequences you're iterating over) is just part of the body.
15:44 BW^-: how would you call arguments to a macro?
15:44 stamourv: I'd call them subforms.
15:44 BW^-: aha
15:45 stamourv: `for/hasheq' contains an implicit begin for its body.
15:45 BW^-: oh really, so these are part of the body?:
15:45 BW^-: (when (eq? (peek-char port) #\,)
15:45 BW^-: (read-char port))
15:45 BW^-: (skip-whitespace port)
15:45 BW^-: (values key value))
15:45 stamourv: Yes.
15:45 BW^-: i.e. rows 19-22
15:45 BW^-: hmm
15:45 BW^-: but it doesn't make sense! )=
15:45 BW^-: =)
15:45 BW^-: so
15:45 BW^-: the input that this procedure takes
15:45 BW^-: is traversing over a JSON hashtable, e.g.
15:45 stamourv: The `values' at the end return the key and the value for the next addition to the hash-table.
15:45 BW^-: { "a": "b", "c": "d" }
15:46 BW^-: stamourv: but thati s already done then no?
15:46 stamourv: The `in-port-until' part does most of the parsing.
15:46 BW^-: stamourv: the values that outputs result into the hashtable is on row 16, no??
15:46 stamourv: And binds the key and the value to `key' and `value'.
15:47 stamourv: No, these are the values that `in-port-until' will produce.
15:47 BW^-: stamourv: right,a nd the return point at row 16 is where the new hashtable element is actually passed into for/hasheq for insertion, no?
15:47 stamourv: These are bound to `key' and `value' on line 4.
15:47 BW^-: hm um
15:47 stamourv: No, that happens on line 22.
15:48 BW^-: really. hmm.
15:48 BW^-: i don't see how this fits together =)
15:48 stamourv: `in-port-until' is a sequence that generates 2 elements at a time.
15:48 BW^-: so what sense does in-port-until make at all?
15:48 BW^-: aha, so it does not actually invoke those two lambdas during the invocation of in-port-until ??
15:48 BW^-: but it just makes like a sequence object for later use??
15:48 stamourv: Its lambda argument tells it how to process its input stream.
15:48 BW^-: that at that time, then, will invoke the lambdas?
15:49 stamourv: Pretty much, yes.
15:49 BW^-: look, so, the input is like this  { "a": "b", "c": "d" }
15:49 BW^-: so, the checking for comma is done at row 19
15:49 BW^-: but thec hecking for } at 18
15:50 BW^-: how does this fit together - i mean, where is the reading actually done?
15:50 stamourv: The checking on line 18 tells `in-port-until' to stop generating elements, which means that `for/hasheq' will stop, since the sequence it iterates over is over.
15:51 stamourv: After that, line 23 makes sure that we did indeed stop because we reached a closing brace (and not because we reached EOF, or something.)
15:51 BW^-: ok
15:51 BW^-: hjm
15:52 BW^-: i think in this port i'm making, i should just rewrite this into a new loop, this is the kind of format i was using anyhow already
15:52 BW^-: the |when| form, is that a synonym for |if|, or does it have properties of a loop?
15:53 stamourv: Re starting from scratch: That's probably a good idea. Gambit doesn't have anything like Racket's `for' loops.
15:53 (join) Kaylin
15:54 BW^-: thanks for the feedback. yeah, just looking at it now i think the logics expressed in this procedure are simple
15:54 stamourv: Re `when': One-armed `if's are forbidden in Racket (and are a terrible idea in general. very easy to make mistakes with them).
15:54 BW^-: i just need to wrap it to use of r5rs procedures only :)
15:54 stamourv: Also, `when' has an implicit `begin'.
15:54 BW^-: aha
15:54 BW^-: aha noted. than you for pointing out.
15:54 stamourv: BW^-: Good luck with that. There's really not much in R5RS.
15:55 BW^-: indeed. :)
15:55 stamourv: I hope you don't mind reimplementing the wheel. ;)
15:55 (quit) jeapostrophe: Ping timeout: 244 seconds
16:00 (join) rich1
16:00 BW^-: stamourv: hehe thanks. well, as far as i know this was never ported to gambit before. i ported this json lib like 4y ago but just ran into this usecase of using the hashtable aspect now
16:00 BW^-: i think i didn't port it originally because i got lost in the for/hasheq form. :)
16:01 BW^-: apart from this, of course code reuse should be maximized, little or no point in reinventing things.
16:01 BW^-: umm
16:01 BW^-: stamourv: are you sure that that code in for/hasheq is the body just ilke that
16:01 BW^-: just look at it
16:01 BW^-: (when* (eq? (peek-char port) #\,)
16:01 BW^-: (read-char port))
16:01 BW^-: (skip-whitespace port)
16:02 BW^-: ^ this indeed reads out a comma section in a json string. but, right after that it ends with:
16:02 BW^-: (expect (read-char port) #\})
16:02 BW^-: which reads out the last character in it.
16:02 BW^-: in for/hasheq's definition, the definition goes something like (for/hasheq test coooond-code .....   lastthing)
16:02 BW^-: sth like this
16:02 BW^-: i didn't get the point there
16:02 soegaard: Apropos one-armed if:  http://stackoverflow.com/q/10863192/23567
16:02 BW^-: though supposedly (expect (read-char port) #\}) is |lastthing|. :)
16:03 dyoo: BW^: just following up no this: are you looking at dherman's json code?
16:03 BW^-: dyoo: yep
16:03 dyoo: If so, you probably want to look at the Racket standard library version instead
16:03 dyoo: my understanding is that the one in the standard library is a descendent of the code in planet
16:03 dyoo: let me see if I can find a uRL for you...
16:03 BW^-: dyoo: aha interesting. what's the URL?
16:03 BW^-: right
16:03 stamourv: BW^-: Re the comma part: That's just skipping the comma, so that the next time we get an element from the sequence, we're looking at an actual element, and not at the comma.
16:03 BW^-: i would guess they're very close in terms of code - this json lib is complete really.
16:04 dyoo: ah, found it: https://github.com/plt/racket/blob/master/collects/json/main.rkt
16:04 stamourv: Re parsing the closing brace: that's not actually in the `for/hasheq', that's in the `begin0'.
16:04 stamourv: Re implementation of `for/hasheq': The last form is special because it's the one that determines what goes in the hash. The others before it are executed for effect.
16:05 (quit) whoops: Quit: Farewell
16:07 BW^-: in PLT, do symbols GC?
16:08 stamourv: What is this PLT of which you speak? ;)
16:08 didi hopes it's food
16:08 didi <- hungry
16:08 stamourv is hungry too.
16:10 dyoo: bw^-: I believe they're usually in an interned table, but that they should be held weakly to be gc-able.
16:10 dyoo: trying to see if the "Inside" internal documentation mentions this...
16:10 (join) whoops
16:11 dyoo: they should be.  They're tagged so that they cooperate with the 3m collector.  There are details in http://docs.racket-lang.org/inside/im_memoryalloc.html#%28part._im~3a3m%29
16:12 (join) jonrafkind
16:12 (join) anRch
16:13 dyoo: BW^-: You could probably experiment by generating temporary random symbols in a loop, and watch if memory usage is bounded.  Given that Racket has support for weak hash tables, it would be funky if symbols couldn't garbage collect.
16:14 cnonim: guys, i'm define syntax parameter, but i create some syntax in runtime, how i can put that syntax parameter to syntax created in runtime?
16:17 dyoo: cnonim: do you have an example on gist.github.com?
16:17 (join) brum
16:17 (join) jbclements
16:18 cnonim: one moment
16:22 (quit) brum: Ping timeout: 248 seconds
16:23 (join) Fare
16:23 cnonim: https://gist.github.com/cNoNim/c6d6b7f91ca63d868401
16:23 (nick) Fare -> Guest92657
16:23 cnonim: how i can put current value of verbosity-level for output "test at runtime"?
16:24 (quit) Guest52818: Ping timeout: 256 seconds
16:25 asumu: cnonim: I'm still confused why you need syntax parameters and not just regular parameters.
16:26 dyoo: this is slightly odd.  I agree with asumu: I expected regular, runtime parameters here.
16:26 cnonim: i can do like https://gist.github.com/cNoNim/525aa84c8ecfc8fc0d5c
16:26 cnonim: but how i can put current value?
16:28 cnonim: if i use regular parameter i got "verbosity-level: unbound identifier in module (in the transformer environment, which does not include the run-time definition)"
16:29 cnonim: rudybot: (define verbosity-level (make-parameter 0))
16:29 rudybot: cnonim: your sandbox is ready
16:29 rudybot: cnonim: Done.
16:29 dyoo: cnonim: https://gist.github.com/dyoo/5060281
16:29 asumu: If you use a regular parameter, you would just include it in the template and you wouldn't unquote it.
16:29 asumu: (yeah, see dyoo's snippet :))
16:30 dyoo: cnonim: if you're doing stuff with eval and you're not writing a dynamic evaluator or repl, you're probably doing something "against the grain".
16:31 cnonim: (define-syntax (log stx) (syntax-case stx () ([_ num msg ...] #`(when (>= (verbosity-level) num) (apply printf (list msg ...))))))
16:31 cnonim: rudybot: (define-syntax (log stx) (syntax-case stx () ([_ num msg ...] #`(when (>= (verbosity-level) num) (apply printf (list msg ...))))))
16:31 rudybot: cnonim: Done.
16:32 cnonim: (log 0 "test")
16:32 cnonim: rudybot: (log 0 "test")
16:32 rudybot: cnonim: ; stdout: "test"
16:32 cnonim: rudybot: (log 1 "test")
16:32 rudybot: cnonim: Done.
16:32 cnonim: rudybot: (parameterize ([verbosity-level 1])(log 1 "test"))
16:32 rudybot: cnonim: ; stdout: "test"
16:33 dyoo: cnonim: does that match what you want then?
16:34 (quit) Kaylin: Quit: Leaving.
16:35 cnonim: one moment
16:37 (quit) jbclements: Quit: jbclements
16:37 (quit) ivan\: Quit: ERC Version 5.3 (IRC client for Emacs)
16:37 cnonim: syntax-parameter have best performance at runtime? or no?
16:37 dyoo: syntax-parameters don't exist at runtime; that's why I'm having difficulty understanding the question
16:38 (join) ivan\
16:38 (join) Guest52818
16:38 dyoo: cnonim: by the way, Racket already has logging infrastructure, so you might just be able to use it:  http://docs.racket-lang.org/reference/logging.html
16:41 cnonim: at runtime #'(when (>= #,(syntax-parameter-value #'verbosity-level) num) 'null) translated to (when (>= 0 1) 'null) and optimizer may be excluded that. i understand correctly?
16:41 cnonim: it's like preprocessor in plain C or C++
16:42 (join) jbclements
16:42 dyoo: Yes, because the instructions are being omitted at compile time.  But that means you can't look at or adjust verbosity level at runtime: it's already been omitted by then.
16:42 cnonim: on this i can write assert macroses and exclude them at production
16:43 cnonim: but on regular parameter I can not do that
16:43 cnonim: this is difference
16:44 dyoo: Yes.  But don't expect to get performance by using eval at runtime: that's almost certainly going to be expensive.
16:45 cnonim: i do not use eval it was just example
16:47 cnonim: I define a syntactic object that defines the syntax object
16:47 cnonim: at runtime
16:48 (quit) dyoo: Quit: dyoo
16:52 (join) bitonic
16:52 (quit) gridaphobe: Remote host closed the connection
16:57 (quit) bitonic: Read error: Operation timed out
17:00 (quit) Licenser: Ping timeout: 260 seconds
17:04 (quit) bremner: Read error: Operation timed out
17:04 (join) gridaphobe
17:06 soegaard: Is this readable? http://rosettacode.org/wiki/Range_expansion#Racket
17:07 cnonim: racket have buffered io?
17:07 soegaard: yes
17:08 soegaard: unless you disable the buffer
17:08 soegaard: http://docs.racket-lang.org/reference/port-buffers.html?q=buffer
17:10 (quit) anRch: Quit: anRch
17:12 (quit) cnonim: Ping timeout: 255 seconds
17:13 (join) bremner
17:13 (quit) bremner: Changing host
17:13 (join) bremner
17:14 (quit) mizu_no_oto: Quit: ["Textual IRC Client: www.textualapp.com"]
17:15 (join) mizu_no_oto
17:18 (join) brum
17:21 (join) Licenser_
17:22 (nick) Licenser_ -> Licenser
17:24 (quit) brum: Ping timeout: 246 seconds
17:27 (join) Nisstyre-laptop
17:28 (quit) basdirks: Ping timeout: 240 seconds
17:29 (quit) mizu_no_oto: Ping timeout: 246 seconds
17:31 asumu: soegaard: looks good to me. (#:match-select was new to me though)
17:31 (join) basdirks
17:34 soegaard: great
17:36 (quit) basdirks: Ping timeout: 255 seconds
17:38 (quit) rich1: Quit: Leaving.
17:44 (join) yours_truly
17:48 greghendershott: I love the Oberon-2 code for it: http://rosettacode.org/wiki/Range_expansion#Oberon-2
17:49 (quit) yours_truly: Quit: Leaving
17:53 soegaard: !?!
17:53 greghendershott: Whereas in APL all that would probably boil down to one line like ⎕←{⍵/⍨~{⍵∨≠\⍵}⍵∊'<>'}txt
17:53 greghendershott: However they'd both take equally long to type out. :)
17:55 soegaard: I miss something shorter than string->number.
18:03 (join) bitonic
18:05 (quit) soegaard: Quit: soegaard
18:15 (quit) dauterive: Quit: Leaving
18:19 (join) brum
18:23 (quit) brum: Ping timeout: 245 seconds
18:24 (quit) gridaphobe: Remote host closed the connection
18:33 (quit) bitonic: Remote host closed the connection
18:33 (join) bitonic
18:37 (join) gridaphobe
18:38 (join) ambrosebs
18:52 (quit) ambrosebs: Remote host closed the connection
18:54 (quit) jonrafkind: Ping timeout: 246 seconds
19:06 greghendershott: soegaard: Well the opposite, number->string, is practically just ~a these days. Thanks to racket/format. That's pretty short. :)
19:11 (quit) SeanTAllen: Ping timeout: 255 seconds
19:12 Nisstyre-laptop: greghendershott: (define number->string ((curry format) "~a"))
19:12 Nisstyre-laptop: although that works for anything really, so idk why anyone would bother
19:19 (join) brum
19:21 (quit) Nisstyre-laptop: Quit: Leaving
19:24 (quit) brum: Ping timeout: 248 seconds
19:32 asumu: So clearly we just need co-~a and we'd be set.
19:36 (quit) didi: Ping timeout: 260 seconds
19:37 (join) yacks
19:37 (join) dnolen
19:40 (quit) gridaphobe: Remote host closed the connection
19:47 (join) mizu_no_oto
19:49 (join) ambrosebs
19:53 (quit) ambrosebs: Remote host closed the connection
19:56 (part) ryan_: "Ex-Chat"
20:09 (join) gridaphobe
20:17 (quit) jbclements: Quit: jbclements
20:18 (join) kvda
20:19 (join) brum
20:22 (quit) dnolen: Ping timeout: 260 seconds
20:24 (quit) brum: Ping timeout: 248 seconds
20:25 (join) brum
20:27 (quit) Guest52818: Ping timeout: 264 seconds
20:34 (quit) mizu_no_oto: Quit: Computer has gone to sleep.
20:36 (join) jeapostrophe
20:52 (quit) carleastlund: Quit: carleastlund
20:57 (join) mizu_no_oto
20:58 (join) yeboot
20:59 yeboot: rudybot: firework
20:59 rudybot: yeboot: (define (firework-launcher) (thing-launcher firework (current-output-port) (random 10)))
20:59 yeboot: rudybot: (exit)
20:59 rudybot: yeboot: your sandbox is ready
20:59 rudybot: yeboot: error: evaluator: terminated (exited)
20:59 yeboot: rudybot: firework
20:59 rudybot: yeboot: your sandbox is ready
20:59 rudybot: yeboot: error: firework: undefined; <NEWLINE>  cannot reference an identifier before its definition <NEWLINE>   in module: 'program
21:00 (quit) brum: Remote host closed the connection
21:03 (quit) cantcode: Quit: ragequit
21:09 (quit) kvda: Quit: z____z
21:09 (join) SeanTAllen
21:13 (join) cantcode2
21:15 (quit) gridaphobe: Remote host closed the connection
21:15 (join) Kaylin
21:23 (join) brum
21:28 yeboot: what did the plt stand for in plt scheme?
21:33 (join) kvda
21:38 jaimef: Pupils Learning To Scheme
21:38 Nisstyre: yeboot: programming language theory
21:39 Nisstyre: http://en.wikipedia.org/wiki/Programming_language_theory
21:39 asumu: PLT doesn't stand for anything.
21:40 Nisstyre: asumu: well, "Programming Language Theory" seems like a valid interpretation at least
21:41 asumu: Yes, but it'd be a backronym.
21:41 Nisstyre: yeah
21:46 (join) RacketCommitBot
21:46 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/t3tnYg
21:46 RacketCommitBot: racket/master 3d0af84 Robby Findler: add macports default path as a place to look for aspell
21:46 (part) RacketCommitBot
21:49 (join) rmathews
21:50 yeboot: rudybot: (define (first-is-last alist) (append (rest alist) (list (first alist))))
21:50 rudybot: yeboot: Done.
21:51 yeboot: rudybot: (first-is-last (list "alpha" "beta" "gamma"))
21:51 rudybot: yeboot: ; Value: ("beta" "gamma" "alpha")
21:51 yeboot: rudybot: (define define 2)
21:51 rudybot: yeboot: Done.
21:52 yeboot: rudybot: (define x 2)
21:52 rudybot: yeboot: error: x: undefined; <NEWLINE>  cannot reference an identifier before its definition <NEWLINE>   in module: 'program
21:52 yeboot: rudybot: (exit)
21:52 rudybot: yeboot: error: evaluator: terminated (exited)
21:53 yeboot: hah, I was about to do (define (exit) (displayln "rosebud") (exit))
22:08 (quit) ijp: Quit: The garbage collector got me
22:33 (quit) bitonic: Ping timeout: 248 seconds
22:38 (join) ambrosebs
22:40 (quit) rmathews: Quit: ...
22:41 (quit) ambrosebs: Remote host closed the connection
22:42 (join) ambrosebs
22:43 (quit) jeapostrophe: Ping timeout: 276 seconds
22:44 (join) Guest52818
22:44 (nick) Guest52818 -> jao
22:44 (join) jao
22:49 yeboot: rudybot: exit
22:49 rudybot: yeboot: your sandbox is ready
22:49 rudybot: yeboot: ; Value: #<procedure:exit>
22:50 yeboot: rudybot: (define (death) (displayln "rosebud") (exit))
22:50 rudybot: yeboot: Done.
22:50 yeboot: (define die death)
22:50 yeboot: rudybot: (define die death)
22:50 rudybot: yeboot: Done.
22:57 (quit) mizu_no_oto: Quit: Computer has gone to sleep.
22:59 (quit) elly: Quit: gone
23:09 yeboot: rudybot: (define die (death))
23:09 rudybot: yeboot: error: evaluator: terminated (exited)
23:09 yeboot: hahah
23:09 yeboot: rudybot: death
23:09 rudybot: yeboot: your sandbox is ready
23:09 rudybot: yeboot: error: death: undefined; <NEWLINE>  cannot reference an identifier before its definition <NEWLINE>   in module: 'program
23:10 yeboot: rudybot: (define (die) (displayln "rosebud") (exit))
23:10 rudybot: yeboot: Done.
23:10 yeboot: rudybot: (die)
23:10 rudybot: yeboot: error: evaluator: terminated (exited)
23:10 yeboot: why won't he print rosebud?
23:19 (join) rmathews
23:33 (quit) yeboot: Read error: Connection reset by peer
23:34 (join) firefux
23:35 asumu: Does anyone know if net/dns has some tests somewhere that I just can't find?
23:35 asumu: (also how does one write a good DNS test anyway? Run a mock DNS server, populate it with test data, and query it?)