00:47 (quit) neilv: Ping timeout: 245 seconds
00:53 (join) tomku
00:54 (quit) rapacity: Remote host closed the connection
00:59 (join) neilv
01:01 (join) rapacity
01:04 (quit) rapacity: Changing host
01:04 (join) rapacity
01:04 (quit) rapacity: Client Quit
01:06 (join) Lajla
01:07 (join) rapacity
01:07 (quit) jeapostrophe: Quit: jeapostrophe
01:56 neilv: if i want to read input using at-exp reader, apply some transformations, and then feed it to scribble for processing like "#lang scribble/doc" does... i can just produce a sexp and do what "scribble/doclang" does, correct? or will my life be easier if i write a .scrbl file that is then re-read by scribble later
02:06 neilv: actually, i think maybe i want to read from input port, do a tiny bit of filtering to produce another input port that at-exp reads from, use syntax-parse or syntax-case on that syntax, and then hand it off scribble/doc processing
02:06 neilv: so scribble has syntax location info to the source document for reporting errors
03:37 (quit) jonrafkind: Ping timeout: 260 seconds
03:49 (join) Demosthenes
03:55 (join) mceier
04:07 eli: neilv: For any post-processing on scribbled contents you can just use it as sexprs.  If syntax properties are messed or not is a different question (which, as usual, raise the question of using a macro or a function instead).
04:19 neilv: eli: thanks. i posted on the list with more question a little earlier. i'm wondering whether planet requires that scribble read from a file. if it does, i have a couple ideas for how to work with that
04:24 eli: neilv: Do you have an example for a source file that you want to convert?
04:25 neilv: very similar to http://planet.racket-lang.org/package-source/neil/html-parsing.plt/1/2/html-parsing.rkt
04:25 rudybot: http://tinyurl.com/4yctpsx
04:27 neilv: scribble-like markup will be on lines starting with ";;;", and the other lines will be read for info pertinent to the documentaiton, such as typed-racket syntax
04:27 eli: neilv: So you want to "massage" it into a scribble file?
04:28 neilv: strictly non-sexually, yes
04:28 eli: And the contents of the doc part will change, right?
04:29 neilv: hm?
04:29 eli: Ah, nevermind, I see that you said that "scribble-like markup will be ...".
04:29 eli: So you want something kind of like the fake-literate-haskell thing, where some lines have the docs, and the rest is the source.
04:30 eli: For now, I think that the best approach would be to hack something dirty that works on whatever level is more convenient.
04:30 neilv: i have been using something dirty for 10 years. i'm ready for it to not be dirty
04:30 neilv: otherwise i can go play video games right now
04:31 eli: An extension that will probably happen soon, is that modules will be able to have more than a single part.
04:31 (quit) ryanaghd1m: Quit: Lost terminal
04:31 eli: This is useful for a bunch of use cases, like having the tests and the code be in the same file, and not load the tests unless you use that part explicitly.
04:31 eli: Another place where it's useful is things like in-line documentation.
04:32 eli: We have two of these currently -- the thing where scribble docs appear on the provide/contract part, and the LP langugae.
04:32 eli: In both cases there's a similar hack in place -- the code itself uses macros that ignores the documentation part.
04:33 eli: (In the LP case it also reorganizes the code according to chunk references, of course.)
04:33 eli: Then, there's a file for the documentation which "extracts" the documentation part.
04:33 eli: So it's kind of like a similar hack, except a little less hacky in the sense that you don't need to generate actual files.
04:34 neilv: i'm trying to do something that works with racket 5.1.3 and doesn't interfere with the #lang line
04:34 eli: The new thing will remove that aspect of it.
04:34 eli: This thing is likely to happen for the next release, which is why I think that a temporary hack is fine.
04:36 neilv: if my doc tool needs to be able to look at ":" and "define:" forms
04:36 neilv: will that work with this module parts thing?
04:36 neilv: or do i still have the same problem of having to read the file myself?
04:36 (quit) realitygrill: Quit: realitygrill
04:37 eli: It will be a little different, similar to the above two.
04:37 eli: Here's a sketch of how it could look:
04:37 eli: you basically replace the ";;; <documentation>" with some real sexpr contents.
04:37 eli: Say, some @doc{blah} parts.
04:37 neilv: if this module parts thing let me add to the language a 'doc' form that is ignored for racket #lang scheme/base  purposes, that would be nice
04:38 eli: Then you write the `doc' macro in a way that makes its contents appear in a doc part of the same module.
04:38 eli: Roughly that -- instead of being ignored, it's made into a kind of a sub-part of the module.
04:38 eli: One that is usually not loaded.
04:39 eli: Unless you're running some code that requires that part explicitly.
04:39 eli: (And that happens when scribbling it.)
04:43 neilv: do we give the info about this documentation language in the #lang line?
04:43 neilv: the information to ignore 'mydoc' forms for purposes of normal loading of the file, at least
04:43 neilv: ?
04:44 eli: I'm not following that completely.
04:45 eli: One thing that you'd have in the lang line is the scribble syntax: #lang at-exp racket-or-whatever
04:45 neilv: #lang at-exp racket/base
04:45 eli: Then, for the ignored `doc' thing, you just get some macro, as usual, from a library.
04:46 eli: The tricky bit would be if you want some code that is also used for the documentation.
04:46 neilv: what else gets added to that so that when we try to treat it as normal racket/base code, it ignores 'mydoc'?
04:46 eli: Nothing.
04:46 eli: Your `mydoc' would be a macro that makes its contents go to the documentation part.
04:46 neilv: where do i say there is this 'mydoc' macro that may be used in this file?
04:47 eli: Huh?
04:47 neilv: in "require"?
04:47 eli: Yeah.
04:48 eli: Same thing as requiring a file now with a `doc' macro that expands to nothing.
04:48 neilv: ok, and that macro expands to some special "(#%part mydoc-part ...)" syntax?
04:48 eli: Yes, that would be the new stuff.
04:48 eli: (Unclear how it'll look in practice though.)
04:49 neilv: ok, so that would be useful for ignoring. but i still have to read the file myself, not say "ok, racket, now expand this module for its 'mydoc-part' part", because ":" and "define:" are in the default part, and not in the "mydoc-part" part?
04:53 eli: That's the tricky bit, which is essentially the same as now.
04:54 eli: You'd need your own definitions that collect information for the docs.
04:54 eli: For example some new define form, which when you use it as some (mydefine (foo x y) ...) will expand into a plain `define', and into some code that is also going into the documentation part.
04:55 neilv: i need to see ":" and "define:", not require people to say "mydefine"
04:57 neilv: i think i will have to stick with reading the file myself for documentation extraction, but i will keep this parts thing in mind so that i can put the docs in sexp form rather than comments
04:59 eli: neilv: There's no need to *actually* call it `mydefine'.  It would obviously be called `define' in my example, or `define:' in yours.
05:03 neilv: aha. before this parts feature is available, i could put the documentation in "#;" comments, with some cookie in them to mark them as documentation
05:07 eli: Well, you can make the macro just ignore them for now, and have a separate post-processor that will un-ignore them as documentation.
05:07 eli: That'd be roughly like those two hacks I mentioned.
05:08 eli: (Where the post-processor doesn't really generate the file, just reads it and creates the documentation directly -- so you have a stub file for the docs that does that and expands into the doc code.)
05:08 neilv: hm. i'd prefer not to need a "require" for ignoring the documentation forms, but i suppose the cost is minor
05:09 eli: That `require' is providing a `doc' binding, which just ignores its contents now, but will later do the other thing.
05:10 neilv: that would also mean that, without parts, i don't need to mess too much with the reader to read the file for documentation purposes (don't need to read comments)
05:10 eli: Yes, that's why it's a good plan -- most of the code will stay.
05:11 neilv: i supposed i could write it in such a way that it's easy to change to parts. that's not how i started writing it
05:12 eli: ...which is why it's a good idea to start right.
05:12 eli: BTW, the stub part is not *that* much of a hack.
05:13 eli: Since it's a macro that reads the original file, it can work with bindings and everything -- so it's even a hygienic kind of tweaking.
05:14 eli: (I'm still not clear on this Irene thing, I stayed the whole night to deal with floods, and there's still nothing...  I thought that it would have been pretty bad by now.)
05:14 neilv: i just this moment pulled up http://www.boston.com/news/nation/specials/110826_irene_winds/?p1=News_links
05:14 rudybot: http://tinyurl.com/4xhcq35
05:15 neilv: it should be around 30mph winds right now
05:17 eli: Certainly doesn't feel like it...
05:17 neilv: so long as i don't get a patio chair through my window, and the power/internet isn't out long, this should be easy (knock on wood)
05:18 eli: The flooding is what worries me, last summer we had about 30cm of water in the first flood.
05:26 neilv: i'm on an upper floor of a red brick elevator building near harvard square. flooding might just mean the tile lobby needs to be waded through til they pump it out
05:36 eli: We have one side of a condo, and our main computing machine is in the basement...
05:47 neilv: tower/desktop, or you have a rack?
05:48 neilv: well, i hope it's fairly uneventful for you.  if your place gets flooded, you guys can crash here for a couple days
05:49 eli: Heh, if it gets bad enough that we need to leave, then you'll have problems too...
05:50 eli: It's just the basement that is prone to flooding...
05:50 eli: (Plain desktop, but some stuff on the floor.)
06:05 ohwow: Hello
06:37 (quit) dsp_: *.net *.split
06:37 (quit) Utkarsh: *.net *.split
06:37 (quit) Lajla: *.net *.split
06:37 (quit) dwaynecrooks: *.net *.split
06:37 (quit) si14: *.net *.split
06:37 (quit) hyko-: *.net *.split
06:37 (quit) rudybot: *.net *.split
06:37 (quit) Demosthenes: *.net *.split
06:37 (quit) DGASAU: *.net *.split
06:37 (quit) shachaf: *.net *.split
06:37 (quit) ve: *.net *.split
06:37 (quit) tewk: *.net *.split
06:37 (quit) stchang_: *.net *.split
06:37 (quit) em: *.net *.split
06:37 (quit) snorble: *.net *.split
06:37 (quit) neilv: *.net *.split
06:37 (quit) jao: *.net *.split
06:37 (quit) tunes: *.net *.split
06:37 (quit) martinhex: *.net *.split
06:37 (quit) wtetzner: *.net *.split
06:37 (quit) samth_away: *.net *.split
06:37 (quit) zakwilson: *.net *.split
06:37 (quit) rapacity: *.net *.split
06:37 (quit) tomku: *.net *.split
06:37 (quit) karljoh: *.net *.split
06:37 (quit) ozzloy: *.net *.split
06:37 (quit) offby1: *.net *.split
06:37 (quit) Fade: *.net *.split
06:37 (quit) svk_: *.net *.split
06:37 (quit) mario-goulart: *.net *.split
06:37 (quit) elliottcable: *.net *.split
06:37 (quit) ohwow: *.net *.split
06:37 (quit) tauntaun: *.net *.split
06:37 (quit) cky: *.net *.split
06:37 (quit) tonyg: *.net *.split
06:37 (quit) mceier: *.net *.split
06:37 (quit) spewn: *.net *.split
06:37 (quit) gf3: *.net *.split
06:37 (quit) Burlingk: *.net *.split
06:38 (quit) janne: *.net *.split
06:38 (quit) stamourv: *.net *.split
06:38 (quit) clklein: *.net *.split
06:38 (quit) ChanServ: *.net *.split
06:38 (join) mceier
06:38 (join) Demosthenes
06:38 (join) rapacity
06:38 (join) Lajla
06:38 (join) neilv
06:38 (join) tomku
06:38 (join) jao
06:38 (join) DGASAU
06:38 (join) karljoh
06:38 (join) Fade
06:38 (join) dwaynecrooks
06:38 (join) tunes
06:38 (join) ChanServ
06:38 (join) tonyg
06:38 (join) gf3
06:38 (join) cky
06:38 (join) tauntaun
06:38 (join) offby1
06:38 (join) clklein
06:38 (join) snorble
06:38 (join) ohwow
06:38 (join) stamourv
06:38 (join) ozzloy
06:38 (join) ve
06:38 (join) janne
06:38 (join) shachaf
06:38 (join) elliottcable
06:38 (join) zakwilson
06:38 (join) em
06:38 (join) mario-goulart
06:38 (join) samth_away
06:38 (join) wtetzner
06:38 (join) martinhex
06:38 (join) rudybot
06:38 (join) hyko-
06:38 (join) stchang_
06:38 (join) tewk
06:38 (join) dsp_
06:38 (join) Utkarsh
06:38 (join) spewn
06:38 (join) svk_
06:38 (join) si14
06:38 (join) Burlingk
06:38 (quit) dsp_: *.net *.split
06:38 (quit) Utkarsh: *.net *.split
06:38 (join) dsp_
06:38 (join) Utkarsh
06:38 (quit) Lajla: *.net *.split
06:38 (quit) dwaynecrooks: *.net *.split
06:38 (quit) si14: *.net *.split
06:38 (quit) hyko-: *.net *.split
06:38 (quit) rudybot: *.net *.split
06:39 (join) Lajla
06:39 (join) dwaynecrooks
06:39 (join) si14
06:39 (join) hyko-
06:39 (join) rudybot
06:39 (quit) Demosthenes: *.net *.split
06:39 (quit) DGASAU: *.net *.split
06:40 (quit) shachaf: *.net *.split
06:40 (quit) ve: *.net *.split
06:40 (join) Demosthenes
06:40 (join) DGASAU
06:40 (join) shachaf
06:40 (join) ve
06:41 (quit) tewk: *.net *.split
06:41 (quit) stchang_: *.net *.split
06:41 (quit) em: *.net *.split
06:41 (quit) snorble: *.net *.split
06:41 (join) tewk
06:41 (join) stchang_
06:41 (join) em
06:41 (join) snorble
06:42 (quit) em: Max SendQ exceeded
06:43 (join) em
06:48 (quit) em: Ping timeout: 252 seconds
06:54 (join) em
06:59 (quit) em: Ping timeout: 245 seconds
07:00 (join) em
07:05 (quit) em: Ping timeout: 246 seconds
07:07 (join) em
07:12 (quit) em: Ping timeout: 264 seconds
07:12 (part) snorble
07:13 (join) em
07:18 (quit) em: Ping timeout: 260 seconds
07:20 (join) em
07:20 (join) MayDaniel
07:24 (join) snorble
07:25 (quit) em: Ping timeout: 260 seconds
07:26 (join) em
07:31 (quit) em: Ping timeout: 245 seconds
07:33 (join) em
07:35 (quit) neilv: Ping timeout: 245 seconds
07:39 (quit) em: Ping timeout: 258 seconds
07:40 (join) em
07:45 (quit) em: Ping timeout: 240 seconds
07:46 (join) em
07:51 (quit) em: Ping timeout: 245 seconds
07:53 (join) em
07:58 (quit) em: Ping timeout: 250 seconds
07:59 (join) em
08:04 (quit) em: Ping timeout: 252 seconds
08:05 (quit) dwaynecrooks: Read error: Connection reset by peer
08:05 (join) em
08:11 (quit) em: Ping timeout: 240 seconds
08:12 (join) em
08:17 (quit) em: Ping timeout: 250 seconds
08:18 (join) em
08:20 (join) dwaynecrooks
08:24 (quit) em: Ping timeout: 260 seconds
08:25 (join) em
08:30 (quit) em: Ping timeout: 258 seconds
08:32 (join) em
08:33 (join) egnarts-ms
08:38 (quit) em: Ping timeout: 276 seconds
08:38 (join) em
08:40 egnarts-ms: could anyone help me with a simple question about Racket: structs, are they built-in types, like lists or vectors, or they are implemented on-top of some basic data structures ?
08:40 egnarts-ms: i've been delving into Racket sources for some time, and was unable to find the answer
08:41 egnarts-ms: neither in the documentation
08:44 (quit) em: Ping timeout: 264 seconds
08:45 (join) em
08:51 (quit) em: Ping timeout: 276 seconds
08:52 (join) em
08:57 (quit) em: Ping timeout: 264 seconds
08:58 (join) em
09:00 (join) jeapostrophe
09:00 (quit) egnarts-ms: Ping timeout: 252 seconds
09:03 (quit) em: Ping timeout: 245 seconds
09:04 (join) em
09:09 (quit) em: Ping timeout: 260 seconds
09:10 (join) em
09:16 (quit) em: Ping timeout: 264 seconds
09:17 (join) em
09:22 (quit) em: Ping timeout: 264 seconds
09:24 (join) em
09:29 (quit) em: Ping timeout: 264 seconds
09:30 (join) em
09:35 (quit) em: Ping timeout: 240 seconds
09:37 (join) em
09:42 (quit) em: Ping timeout: 252 seconds
09:43 (join) em
09:48 (quit) em: Ping timeout: 258 seconds
09:49 (join) em
09:49 (quit) em: Changing host
09:49 (join) em
09:54 (join) RacketCommitBot
09:54 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/ZZTrNQ
09:54 RacketCommitBot: [racket/master] remove leftover tmp file during reference build - Matthew Flatt
09:54 (part) RacketCommitBot
09:55 (quit) em: Ping timeout: 252 seconds
09:56 (join) em
10:01 (quit) em: Ping timeout: 246 seconds
10:01 (join) egnarts-ms
10:02 (join) em
10:07 (quit) em: Ping timeout: 264 seconds
10:08 (quit) jeapostrophe: Quit: jeapostrophe
10:08 (join) em
10:14 (quit) em: Ping timeout: 276 seconds
10:14 (join) em
10:19 ohwow: git.io, that's a cool new domain for github :]
10:19 (quit) em: Ping timeout: 250 seconds
10:21 (join) em
10:25 (quit) em: Ping timeout: 252 seconds
10:27 (join) em
10:32 (quit) em: Ping timeout: 246 seconds
10:33 (join) em
10:38 (quit) em: Ping timeout: 250 seconds
10:39 (join) em
10:40 (join) anRch
10:44 (quit) em: Ping timeout: 240 seconds
10:44 (join) jeapostrophe
10:45 (join) em
10:47 (quit) jeapostrophe: Client Quit
10:51 (quit) em: Ping timeout: 260 seconds
10:52 (join) em
10:57 (quit) em: Ping timeout: 240 seconds
10:57 (quit) anRch: Quit: anRch
10:58 (join) em
11:03 (quit) em: Ping timeout: 250 seconds
11:04 (join) em
11:10 (quit) em: Ping timeout: 264 seconds
11:10 (join) em
11:16 (quit) em: Ping timeout: 252 seconds
11:17 (join) em
11:21 (join) jonrafkind
11:23 (quit) em: Ping timeout: 246 seconds
11:28 (join) em
11:40 (quit) egnarts-ms: Ping timeout: 252 seconds
12:09 (join) samth
12:31 (join) dnolen
12:40 (quit) samth: Ping timeout: 245 seconds
13:05 (join) otzi
13:25 (join) RacketCommitBot
13:26 RacketCommitBot: [racket] plt pushed 3 new commits to master: http://git.io/72EVHw
13:26 RacketCommitBot: [racket/master] allow phase 1+ references to not-yet-defined bindings - Matthew Flatt
13:26 RacketCommitBot: [racket/master] fix cross-reference - Matthew Flatt
13:26 RacketCommitBot: [racket/master] make bytecode compiler less picky about exports - Matthew Flatt
13:26 (part) RacketCommitBot
13:31 (join) RacketCommitBot
13:31 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/qwYZEA
13:31 RacketCommitBot: [racket/master] adjust the online check syntax so that you get a subtle indication of read errors instead of just silence - Robby Findler
13:31 (part) RacketCommitBot
14:14 (quit) tomku: Ping timeout: 252 seconds
14:16 (join) tomku
14:24 (join) ryanaghd1m
15:14 (join) jeapostrophe
16:15 (quit) jeapostrophe: Quit: jeapostrophe
16:15 (join) samth
16:26 (join) jeapostrophe
16:29 (quit) jeapostrophe: Client Quit
16:29 (join) jeapostrophe
16:29 (quit) jeapostrophe: Client Quit
16:31 (join) jeapostrophe
16:32 (quit) jeapostrophe: Client Quit
16:44 (quit) ryanaghd1m: Ping timeout: 260 seconds
16:45 (join) jeapostrophe
16:45 (quit) jeapostrophe: Client Quit
16:51 (join) realitygrill
17:26 (quit) samth: Ping timeout: 245 seconds
17:40 (join) jeapostrophe
17:40 (quit) jeapostrophe: Client Quit
17:42 (quit) mceier: Quit: leaving
18:15 (join) jeapostrophe
18:19 (quit) DGASAU: Remote host closed the connection
18:19 (join) DGASAU
18:23 (join) ryanaghd1m
18:25 (quit) ryanaghd1m: Client Quit
18:26 (join) ryanaghd1m
18:36 (quit) jeapostrophe: Quit: jeapostrophe
18:42 (quit) MayDaniel: Read error: Connection reset by peer
19:03 (join) jeapostrophe
19:22 (quit) jeapostrophe: Quit: jeapostrophe
19:29 (quit) Burlingk: Quit: Leaving
19:40 (join) jeapostrophe
19:42 (quit) jeapostrophe: Client Quit
19:55 (nick) otzi -> flaggy
20:16 (join) RacketCommitBot
20:16 RacketCommitBot: [racket] plt pushed 2 new commits to master: http://git.io/WIFxLw
20:16 RacketCommitBot: [racket/master] move Reference reader & prnter docs to the model chapel - Matthew Flatt
20:16 RacketCommitBot: [racket/master] restore module cache - Matthew Flatt
20:16 (part) RacketCommitBot
20:16 (join) jeapostrophe
20:48 (quit) ryanaghd1m: Ping timeout: 240 seconds
21:18 (join) neilv
21:24 (join) rlph
21:25 (part) rlph
21:37 (join) RacketCommitBot
21:37 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/0Tqt1A
21:37 RacketCommitBot: [racket/master] path normalization in setup/scribble - Matthew Flatt
21:37 (part) RacketCommitBot
21:54 (join) ryanaghd1m
21:58 (join) crk
21:58 crk: hi, i'm facing a strange issue
21:58 crk: i've got this file "insertionsort.rkt" which has a (define (insertionsort ...))
21:58 crk: i run mzscheme in the same directory
21:58 crk: followed by
21:59 crk: > (load "insertionsort.rkt")
21:59 crk: but the insertionsort function does not load
21:59 crk: there's no error in the load
21:59 jonrafkind: how do you know it doesnt load?
21:59 crk: > insertionsort
21:59 crk: reference to undefined identifier: insertionsort
21:59 crk: === context ===
21:59 crk: /usr/lib/racket/collects/racket/private/misc.rkt:85:7
21:59 jonrafkind: you mena you have typed literally (define (insertionsort ...)) ?
21:59 crk: thats how :-|
21:59 crk: yeah...
21:59 crk: its in the file like that
22:00 crk: ideally i would've expected it to say procedure something something
22:00 crk: > number?
22:00 crk: #<procedure:number?>
22:00 jonrafkind: does it say antyhing after you do (load ...) ?
22:00 crk: am I doing something wrong?
22:00 crk: nope...
22:00 crk: I get the REPL prompt again
22:00 crk: no errors/warnings
22:00 crk: using racket 5.1.3
22:00 jonrafkind: i dunno, paste the contents of insertionsort.rkt somewhere (not here)
22:01 crk: yeah
22:01 crk: pastebins :)
22:01 crk: hang on
22:01 crk: http://fpaste.org/0tDG/
22:02 jonrafkind: it may be because of the #lang ...
22:03 jonrafkind: ok do this: in insertionsort.rkt add (provide insertionsort)
22:03 crk: ok
22:03 jonrafkind: then use (require "insertionsort.rkt") instead of (load ...)
22:03 crk: remove the #lang ?
22:03 crk: or let it stay
22:03 jonrafkind: you can let it stay
22:03 jonrafkind: if you leave it there then you have a module and you have to use provide and (require)
22:03 jonrafkind: if you take it out then you can just use (load)
22:03 crk: i see
22:03 crk: wait, let me try that
22:04 crk: adding the line isnt helping
22:04 crk: let me remove #lang and see
22:04 crk: oh dang... wait
22:04 crk: didnt use require :)
22:04 crk is a dork
22:04 (join) RacketCommitBot
22:04 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/ZIcwJA
22:04 RacketCommitBot: [racket/master] fix `compile-omit-files' handling for Planet packages - Matthew Flatt
22:04 (part) RacketCommitBot
22:05 crk: Dr. Flatt... the interview on FLOSS weekly was quite interesting :)
22:05 jonrafkind: so if you want to use modules then the only things you can use outside the module are things you give to (provide ...)
22:06 crk: ok
22:06 crk: it works now :)
22:06 jonrafkind: unless you use (enter!) but thats another barrel of fish
22:06 crk: so if I use a (provide ) in the file
22:06 crk: it makes the file a module, implicitly
22:06 crk: no (module ... )
22:06 crk: ?
22:06 jonrafkind: no if you use #lang racket, then its a module
22:06 jonrafkind: and you load modules with (require ...)
22:06 crk: ok
22:07 jonrafkind: and you get access to things in the module by putting (provide foobar) inside the module
22:07 crk: great
22:07 crk: :)
22:07 jonrafkind: and of course modules can load each other, which is the whole point
22:07 crk: on a side-note, my insertionsort
22:07 crk: which i gave you the link for
22:07 jonrafkind: if you aren't doing anything fancy you can do (load "whatever.rkt") in your files as well and it will usually work
22:07 crk: seems rather procedural, with set! and all that
22:07 crk: i mean, it shows i'm basically a python guy :P
22:08 crk: using the same insert, is there a better/more elegant/scheme way
22:08 jonrafkind: you can use for/fold there
22:08 jonrafkind: i will leave it as an exercise for you to figure out :)
22:08 crk: :D
22:08 crk: you sound like a professor in your own right
22:08 jonrafkind: but in general if you can only do things with mutation thats fine
22:08 crk: but thanks for the help
22:08 jonrafkind: yea rofl
22:09 crk: hmm
22:09 crk: actually this version takes double the amount of space
22:09 jonrafkind: i have to use phrases like "in particular" and "more generally" to become a professor ;p
22:09 crk: i was hoping to make an in-place version
22:09 (quit) jeapostrophe: Quit: jeapostrophe
22:09 crk: I even have the idea in mind... but my scheme-think isnt that evolved, I guess :)
22:10 crk: but as you said, exercise it is
22:10 jonrafkind: also replace (equal? 0 (length l)) with (null? l)
22:10 jonrafkind: length will compute the length of the list each time, so thats O(N), whereas (null? l) is O(1)
22:11 crk: cool
22:11 crk: doing that rightaway
22:42 (join) bayildi
22:42 (part) bayildi
22:47 (quit) crk: Quit: Leaving
22:49 (quit) ryanaghd1m: Ping timeout: 246 seconds
22:50 (quit) jao: Ping timeout: 245 seconds
22:57 (join) jeapostrophe
22:57 eli: jonrafkind: First advice to questions like that should be to dump `load'.
22:58 eli: (And using `mzscheme' as the executable is another problem, of course, giving him the `mzscheme' language bindings.)
22:58 jonrafkind: psh, whatever gets him to a working system > doing things the right way
22:59 (quit) jeapostrophe: Client Quit
23:00 eli: That is an unimaginably wrong answer.
23:00 eli: You could just have told him to (eval (read (open-input-string "(require \"whatever.rkt\")")))
23:00 eli: That would also get him a working system.
23:02 jonrafkind: so what.. if it ever fails he can figure out the right way later
23:02 neilv: is it april first already?
23:03 eli: You disserviced by letting him go on using `load', and that will inevitably make him go into more traps.
23:03 jonrafkind: you do me a disservice by not reading the exchange.. he ultimately used `require'
23:03 eli: You *are* in a CS dept, you should therefore appreciate the value of educating someone over getting them a "working system".
23:03 eli: I know he did.
23:03 neilv: these need not be mutually exclusive goals
23:03 eli: But your first response was bad.
23:04 jonrafkind: i have worked with dozens of systems i barely understand and my frustration reaches epic proportions when I am told to do things "the right way" when I just want to get something done, so i have sympathy for such people
23:04 neilv: you two guys are two ends of the spectrum
23:04 jonrafkind nods
23:05 eli: I have worked with dozen times more dozens of systems.  You get no excuse for just "hacking things".
23:05 jonrafkind: yea right, like im going to read about autoconf when im trying to get some stupid system to build. of course im just going to hack it until it works
23:06 eli: On a larger scale "just getting a working system" is the exact antithesis HtDP.  As a student of one of its authors, it's bad that you follow that.
23:06 eli: (It's also the embodiment of the dot-com bubble burst, and other related bubbles.)
23:07 jonrafkind: straw man argument
23:07 (join) ryanaghd1m
23:07 eli: WTF is the straw here?
23:08 jonrafkind: to get anything working its always reasonable to hack it
23:08 eli: And this is the precise point that HtDP goes agaist.  No straws here.
23:10 jonrafkind: i dont want this situation: tell user to use require/provide, he fools around for 5 hours,  fails, gives up. id rather this situation: user uses load, does some stuff, eventually runs into subtle errors, asks for help, is told to use require, is finally motivated to get past the troubles of require/provide to get it working
23:11 jonrafkind: that is, learning by experience >*infinity learning by being told
23:12 eli: The 5 hour time spendage is a valid argument; I'd argue that you should have helped him to avoid the 5 hours; unfortunately (for you) you validated my point by the actual exchange.
23:12 jonrafkind: if he understands (load) id rather just help him with that than to understand N other concepts (since thats a lot of work for me)
23:14 eli: Anyone who really *understands* `load' and using it in Racket would not have such questions.  It's therefore possible to assume that such askers do not understand it at all.
23:14 jonrafkind: thats true, but since hes really not doing anything complex whatsever, theres almost no harm in using `load' anyway
23:14 jonrafkind: modules are a solution for a problem he will likely never have
23:15 eli sighs
23:15 jonrafkind: and if he ever has the problem he can learn about it then
23:15 eli: Please stop this nonsense.
23:15 jonrafkind: do you think that he will use load, think "hey load is great why should I use require" and then never learn to use require?
23:16 eli: Yes, and I have decades of Scheme history to back that up.  All the way up to R7RS discussions.
23:19 jonrafkind: i just assume people will learn the truth eventually once they have enough practical experience to understand the concepts involved. preaching advanced techniques too early will just confuse people
23:21 eli: (a) there's nothing advanced in `require', if anything, it's `load' that is more advanced; (b) you're dealing with a *ton* of Scheme related material that encourages people to use `load' as the way to structure code -- all stuff that has changed only very recently, which is why `load' will continue to be a problem for a long time.
23:21 jonrafkind: the advanced thing is "why do I need provide"
23:22 eli: For that there's `enter!' (which you specifically discouraged him away from), or -- much better -- use drr.
23:22 jonrafkind: hes from python land so he knows what "import" does of which `load' is basically the same thing
23:23 jonrafkind: i mean, why aren't you berating me for telling him mutation is ok?
23:23 eli: I hightly doubt that Python's import is an open+read+eval thing.  Last time I checked they didn't even get there eval straight.
23:23 eli: Mutation is a much less fundamental problem.
23:23 jonrafkind: the similarity is open a file, get access to all its contents somehow
23:24 eli: And as a language that does provide mutations, we *do* go the practical route of making it possible.
23:24 eli: By "open a file and get access to contents" you're not saying much.  The `eval' that is implicit in `load' is the problem.
23:51 (join) jeapostrophe