00:07 (join) littlebobby
00:11 (quit) samth_: Read error: Connection reset by peer
00:57 (quit) dnolen: Quit: dnolen
01:00 (join) jonrafkind
01:14 (quit) littlebobby: Quit: Ex-Chat
01:24 (quit) ckrailo: Quit: Computer has gone to sleep.
01:27 neilv: eli: another argument in favor of doing things the meow way rather than the old emacs way :) http://www.neilvandyke.org/weblog/2011/06/#2011-06-20
01:31 Nafai: neilv: Will meow support all of the standard racket languages, like typed racket?
01:33 neilv: Nafai: ones using the normal racket reader, yes
01:33 neilv: i will probably also eventually add support for the scribble reader
01:36 (join) mye
01:36 eli: neilv: I'm not sure it's an argument against "the emacs way" -- my guess is that the actual implementation of #; in newer Emacs is just broken.
01:37 eli: That is, I'm 90% sure that there's a way to do it better -- the main problem of what it currently does is that it actually makes the text be considered as a comment, which is fundamentally broken with #;s.
01:37 eli: (Haven't tried fixing it though.)
01:39 neilv: i think i'll implement at least some rudimentary indenting and then do an alpha release
01:49 (quit) jonrafkind: Ping timeout: 276 seconds
01:54 sheikra: hey, is there a way to format floating point number to a certain digit?
01:55 sheikra: printf and format functions don't seem to have a format string that can do this
01:56 neilv: http://docs.racket-lang.org/reference/generic-numbers.html?q=real&q=number#(def._((lib._racket/private/base..rkt)._real-~3edecimal-string))
01:56 rudybot: http://tinyurl.com/5vhea2z
02:00 sheikra: neilv: I guess that's real->decimal-string?
02:00 neilv: yes
02:01 sheikra: thanks
02:03 (join) gienah
02:32 (quit) mye: Remote host closed the connection
02:33 (join) mye
02:41 (join) hkBst
02:42 (quit) hkBst: Client Quit
02:43 (join) hkBst
02:46 (join) andrew_
02:47 andrew_: Hallo
03:20 (quit) andrew_: Ping timeout: 252 seconds
03:52 (quit) realitygrill: Quit: realitygrill
03:53 (join) masm
04:02 (quit) sheikra: Quit: Leaving
04:05 (quit) rgee: Ping timeout: 244 seconds
04:14 (quit) neilv: Quit: Leaving
04:51 (join) misterncw
04:57 (quit) mye: Ping timeout: 276 seconds
07:31 (join) samth_
08:13 (join) dnolen
08:48 (join) racket
08:48 racket: [racket] plt pushed 1 new commit to master: http://bit.ly/mm0etV
08:48 racket: [racket/master] document `scribble' command-line use - Matthew Flatt
08:48 (part) racket
09:05 (join) MayDaniel
09:19 (join) mithos28
09:24 (quit) samth_: Ping timeout: 250 seconds
09:39 (quit) mithos28: Quit: mithos28
09:51 (part) Nafai: "WeeChat 0.3.5"
10:00 (join) rgee
10:00 (nick) bremner_ -> bremner
10:08 (join) mye
10:18 (join) Skola
10:18 Skola: hello there!
10:18 Skola: I have a Haskell/Python/JS background and want to get going with Racket
10:19 offby1: skoal
10:19 offby1: Skola: with that experience, you should find it pretty straightforward
10:19 Skola: alright
10:19 Skola: need something to get me going, could you recommend a tutorial or guide?
10:20 Skola: like not-too-slow
10:20 offby1: well, there's a bunch of 'em at http://docs.racket-lang.org/index.html
10:20 offby1: "not too slow" might be http://docs.racket-lang.org/continue/index.html
10:21 Skola: alright, thanks
10:24 (quit) MayDaniel: Read error: Connection reset by peer
10:25 (quit) dnolen: Quit: dnolen
10:41 (quit) Skola: Ping timeout: 252 seconds
10:43 (quit) gienah: Quit: leaving
10:46 (join) Skola
10:54 (join) bugQ
10:56 (join) mithos28
11:05 (quit) hkBst: Read error: Connection reset by peer
11:10 (quit) bugQ: Ping timeout: 264 seconds
11:14 (join) realitygrill
11:16 (join) bugQ
11:21 (nick) elliottcable -> ec|detached
11:28 (join) dnolen
11:33 (quit) activedecay: Read error: Connection reset by peer
11:35 (quit) bugQ: Ping timeout: 240 seconds
11:35 (join) ckrailo
11:36 (join) anRch
11:38 (nick) ec|detached -> elliottcable
12:08 (nick) samth_away -> samth
12:10 (quit) anRch: Quit: anRch
12:11 (join) MayDaniel
12:11 (join) karljoh
12:18 (join) jonrafkind
12:21 clklein: Is there a better tool than `section' to build an appendix for a paper written in Scribble?
12:22 clklein: The problem with my current setup is that I can't force a page break before the appendix.
12:23 clklein: The appendix `section' is after the bibliography (generated by define-cite). The bibliography seems to be eating my @(element (style "newpage" '()) '())
12:24 clklein: i.e., I have @(generate-bibliography) @(element (style "newpage" '()) '()) @section[#:tag "sec:proof"]{Appendix}
12:24 clklein: but the \newpage doesn't appear in the generated latex
12:26 (quit) misterncw: Remote host closed the connection
12:26 samth: clklein: you might need something more heavyweight to make it appear
12:26 samth: i use something like this:
12:27 samth: (define (exact . items)
12:27 samth: (make-element (make-style "identity" '(exact-chars))
12:27 samth: items))
12:27 samth: and then if you have macro named \identity
12:27 samth: you can dump whatever latex you want
12:27 samth: although you should be able to just change that function to output "newpage" directly
12:28 clklein: But then the only difference is the exact-chars. Could that matter?
12:28 clklein: I'll check.
12:29 samth: clklein: oh, by the way: https://github.com/samth/var/blob/master/redex-notes.txt (in case you run out of other things to do :)
12:30 samth: david van horn and i have been collecting things we want from redex
12:32 clklein: nice
12:32 samth: they vary a lot in difficulty
12:32 clklein: yes definitely :)
12:33 (join) anRch
12:33 samth: did 'exact-chars work?
12:34 clklein: Oh, sorry, I got side-tracked by your list. No, it didn't.
12:34 samth: darn
12:35 samth: try the identity trick, i'd say
12:35 samth: oh, and if you need explanations of the items in the list, i'm happy to provide
12:36 samth: and, of course, the whole thing we're doing would be totally impossible w/o redex, so thanks for that
12:36 clklein: OK, thanks. I may do some of the easy ones this week, and I'm planning to start on the whole "run FASTER" thing after the POPL deadline.
12:37 samth: honestly, running faster isn't as important as most of the others, i think
12:38 clklein: It's never been a problem for me either, but we need it if we want to do models of realistic languages
12:39 clklein: For example, Redex's speed forced Arjun to write an interpreter for \JS
12:39 samth: it would be nice, but i think ultimately real tools require writing real tools
12:40 (join) ckrailo_
12:40 samth: although \js is an interesting middle ground
12:40 clklein: We're not after a real implementation -- just one that's competitive with a naive interpreter
12:41 (quit) anRch: Read error: Connection reset by peer
12:41 clklein: i.e., I'm talking about something much more modest than that whole compiler-generator idea
12:41 (join) anRch
12:42 (quit) ckrailo: Ping timeout: 246 seconds
12:42 (nick) ckrailo_ -> ckrailo
12:43 ohwow: ocaml can be compiled to native code without adding runtime library iirc
12:43 samth: clklein: i understand
12:43 samth: ohwow: that's not correct
12:43 ohwow: hm i think i wrote this to wrong channel
12:44 ohwow: samth: so every ocaml compiler adds its' runtime?
12:44 samth: ohwow: there's only one ocaml native compiler, and the binaries it produces link to the runtime
12:45 clklein: samth: Jay and some folks at BYU also built a model (some kind of concurrency thing IIRC) that ended up being prohibitively slow for reasons that he thinks Redex could tackle.
12:45 ohwow: I think I am mistaking OCaml for SML than.
12:45 samth: that's how they use a garbage collector
12:45 samth: this is true of SML as well, and of C
12:45 ohwow: But Racket and Chicken are pretty fast tho
12:46 ohwow: some people claim that "C is the fastest language", but I think it depends on the compiler
12:46 ohwow: and on algorithms used, of course
12:46 samth: yes, although in general similar programs in C and Racket will run faster in C
12:46 samth: clklein: i remember jay talking about that
12:47 samth: i think the most interesting idea in the list is the disappearing traces
12:47 samth: which is something people do in the pi calculus, among other places
12:48 clklein: samth: I don't know what you mean. Are you imagining some kind of interactive way to hide traces?
12:49 samth: no, i'm imagining that if you reduce to a particular point, you can tell redex that this whole branch of the tree didn't happen
12:49 (quit) anRch: Ping timeout: 255 seconds
12:49 samth: and it would all disappear from redex's point of view
12:50 samth: back to the last step with multiple successors
12:50 clklein: oh
12:50 clklein: How do you specify that in a paper?
12:51 clklein: Is there some standard notation or do you just say so in the prose?
12:52 samth: i'm not sure what the notation is
12:52 samth: but i'll see what i can find
12:52 clklein: thanks
12:53 (join) ckrailo_
12:55 clklein: samth: @(make-element (make-style "identity" '(exact-chars)) '("stuff")) doesn't show up in the latex either
12:55 (quit) ckrailo: Ping timeout: 260 seconds
12:55 (nick) ckrailo_ -> ckrailo
12:55 samth: clklein: something larger must be going wrong there
12:56 clklein: I blame define-cite's bibliography generator.
12:58 (join) ckrailo_
12:59 (quit) mithos28: Quit: mithos28
13:00 (quit) ckrailo: Ping timeout: 252 seconds
13:00 (nick) ckrailo_ -> ckrailo
13:13 (join) mithos28
13:14 (join) mithos28_
13:14 (quit) mithos28: Read error: Connection reset by peer
13:14 (nick) mithos28_ -> mithos28
13:14 (join) anRch
13:14 samth: clklein: apparently the the standard thing is to have an "assume" construct
13:15 samth: where if the assumption is invalidated, then this trace didn't happen
13:15 samth: although this is usually presented in non-operational models of language semantics
13:15 samth: here's an example: http://research.microsoft.com/en-us/um/people/leino/papers/krml157.pdf
13:16 samth: although that doesn't really have a normal semantics in it
13:23 clklein: samth: OK, I'll take a look.
13:23 samth: clklein: when you're here, i'll make you talk to aaron, who i learned this stuff from
13:24 clklein: Is this something you're currently doing by writing something like apply-reduction-relation* that does something like their `assume'
13:24 clklein: sounds good
13:24 samth: no, this is something where i'm just currently pretending that those traces don't exist
13:28 clklein: when you look at the traces graph?
13:28 samth: yeah
13:28 clklein: Ah, I see why you need tool support then :)
13:31 (join) bugQ
13:33 (join) tauntaun
13:39 (quit) anRch: Ping timeout: 276 seconds
13:40 (join) anRch
13:40 (quit) bugQ: Ping timeout: 240 seconds
13:46 (join) bugQ
13:54 (join) dyoo
14:26 (join) redgetan
14:56 (quit) redgetan: Ping timeout: 240 seconds
15:00 (join) redgetan
15:06 (quit) anRch: Quit: anRch
15:12 samth: mithos28: merging your bugfix now
15:14 mithos28: samth: ok, and I'll be pushing the rest of the tests for the big pull request in a minute
15:15 samth: great
15:15 mithos28: That is unless you think I missed testing anything (after looking at them)
15:18 mithos28: pushed
15:32 (join) racket
15:32 racket: [racket] plt pushed 1 new commit to master: http://bit.ly/iW83ZM
15:32 racket: [racket/master] Clarifying function contracts for s/s/d callers - Jay McCarthy
15:32 (part) racket
15:35 samth: mithos28: pushed
15:35 (join) racket
15:35 racket: [racket] plt pushed 5 new commits to master: http://bit.ly/lEYxrR
15:35 racket: [racket/master] Fixes TR regexp regression. Closes PR11991. - Eric Dobson
15:35 racket: [racket/master] Include `typed-scheme' language in doc check. - Sam Tobin-Hochstadt
15:35 racket: [racket/master] Support additional `provide'. - Sam Tobin-Hochstadt
15:35 (part) racket
15:44 (quit) rapacity: Ping timeout: 258 seconds
15:45 (join) Utkarsh_
15:45 (quit) penryu: Ping timeout: 246 seconds
15:46 (quit) Utkarsh: Ping timeout: 276 seconds
15:49 (join) Utkarsh
15:50 (quit) dyoo: Ping timeout: 252 seconds
15:50 (quit) Utkarsh_: Ping timeout: 264 seconds
15:51 (join) penryu
15:51 (join) rapacity
15:55 (nick) elliottcable -> ec|detached
15:57 (nick) ec|detached -> elliottcable
16:02 (nick) elliottcable -> elliottcabal
16:02 (quit) redgetan: Read error: Connection reset by peer
16:26 (join) jao
16:29 (quit) tauntaun: Quit: Ex-Chat
16:30 (join) activedecay
16:35 (quit) bugQ: Ping timeout: 244 seconds
16:38 (nick) samth -> samth_away
16:46 (nick) elliottcabal -> elliottcable
17:03 (quit) MayDaniel: Read error: Connection reset by peer
17:06 mithos28: stamourv or samth_away: ping
17:06 stamourv: mithos28: pong
17:07 mithos28: Is there a way when TR says that it has insufficient information to type check to see what the problem is
17:09 stamourv: That message happens only when the variable it can't type is not found in the original program.
17:09 stamourv: i.e. has been introdued by expansion.
17:09 stamourv: The error used to give the identifier, but since it only exists in the expansion, that was somewhat confusing for users, so I got rid of it.
17:10 mithos28: Ok, I'll run an older version to get the untyped id.
17:10 stamourv: The classic case would be a use of the `for' macros without enough type annotations. In these cases, the typechecker would complain about the `for-loop' variable being untyped. That variable is introduced by expansion.
17:11 stamourv: Or, you could change the error message to include the identifier, for debugging.
17:11 stamourv: What are you trying to do?
17:11 mithos28: The use case I need it for is typing an untyped library, and I am writing a test program, which uses a macro expanding into an untyped id.
17:12 stamourv: The message is in private/type-annotation.rkt, line 105.
17:12 stamourv: Do you think that seeing the untyped id could be useful for users in general?
17:13 mithos28: no, I always got confused by it with the for loops
17:13 stamourv: If so, we could add it to the current message, but that may be confusing.
17:13 stamourv: Ok.
17:14 stamourv: Also, the untyped id is only omitted if it's not in the original source. Otherwise, it's displayed in the error.
17:14 mithos28: If the macro expands to a definition, do I pretty much need to rewrite the macro to make it type check
17:14 stamourv: It depends.
17:15 stamourv: You can probably get away with writing your own wrapper macro that attaches annotations to identifiers before calling out to the original macro.
17:15 stamourv: That's what the `for:'
17:15 stamourv: macros do.
17:15 stamourv: Reimplementing `for' would have been a _bad_ idea.
17:16 mithos28: but the definition in this case is entirely internal to the macro
17:20 stamourv: In some cases, adding an expected type (with a type annotation on the whole macro invocation) is enough to nudge inference in the right direction.
17:21 stamourv: Otherwise, yes, it seems like you'd have to reimplement the macro.
17:21 stamourv: Interaction between TR and macros is one of the big open problems we currently have.
17:21 stamourv: Of course, having full type inference would fix that, but I'm not sure that's possible.
17:22 stamourv: By the way, which macro is causing you trouble?
17:22 mithos28: test-suite
17:22 mithos28: from rackunit
17:22 mithos28: and I found a bug with which syntax object TR blames, which I'll submit in a sec
17:22 stamourv: typed/rackunit doesn't provide it?
17:24 mithos28: I'm fixing up typed/rackunit, and it just reprovides it
17:25 stamourv: As in, it reprovides it untyped?
17:26 mithos28: as in (require (only-in rackunit test-suite)) (provide test-suite)
17:27 stamourv: Ah.
17:31 mithos28: Do you see an option besides rewriting it/ changing it so that the expansion works with TR?
17:36 stamourv: Well, you could have a wrapper macro that local-expands it, walks the expansion and adds annotations to the right places, but that's a terrible idea.
17:36 stamourv: So, the answer is: not really.
17:37 stamourv: As I said, that's one of the problems that we currently don't know how to fix.
17:37 mithos28: ok. Well I'll just use with-typed to write my tests then, and use the untyped worlds test-suite
17:47 (join) MayDaniel
17:53 (quit) Skola: Ping timeout: 240 seconds
17:53 (join) Skola
18:14 (quit) dnolen: Quit: Page closed
18:17 jonrafkind: given some .rkt file how can I compile it to a .zo programmatically?
18:17 jonrafkind: (without invkoing raco make on the command line or some such thing)
18:19 (quit) activedecay: Quit: Changing server
18:22 (quit) mye: Ping timeout: 246 seconds
18:25 (quit) MayDaniel: Read error: Connection reset by peer
18:26 stamourv: jonrafkind: (require compiler/compiler)
18:26 stamourv: compile-zos is probably what you want.
18:28 jonrafkind: ok so i should probably call (compile-zos (make-base-namespace)) ?
18:29 stamourv: If you want to compile module files, call it with #f.
18:30 stamourv: To compile files with toplevel interactions, use (void) instead.
18:30 stamourv: 5th paragraph in compiler/compiler's docs.
18:31 mithos28: starmourv: If I find TR bugs that should be closed (as in already been fixed), how should I report them?
18:31 jonrafkind: ok i dont see why I would want to used the current namespace to compile modules, but whatever it seems to work
18:34 stamourv: mithos28: Email me the numbers, and I'll close them. Preferably in one big batch.
18:35 mithos28: stamourv: ok
18:35 stamourv: jonrafkind: I assume it compiles modules with a fresh state, so whatever you give it is dropped.
18:35 jonrafkind: sounds likely
18:36 jonrafkind: oh module? should be #t
18:36 jonrafkind: wtf..
18:46 (quit) sethalves: Ping timeout: 246 seconds
19:02 (quit) mithos28: Quit: mithos28
19:11 (quit) realitygrill: Quit: realitygrill
19:18 (join) dnolen
19:26 (join) Nightwolf
19:32 (join) mithos28
19:33 (quit) Skola: Ping timeout: 264 seconds
19:47 (quit) mithos28: Quit: mithos28
19:53 (join) mithos28
19:58 (quit) jao: Ping timeout: 255 seconds
20:03 (quit) janne: Ping timeout: 258 seconds
20:04 (join) janne
20:09 (quit) mithos28: Quit: mithos28
20:13 (join) mithos28
20:18 (quit) ckrailo: Quit: Computer has gone to sleep.
20:28 (quit) masm: Ping timeout: 240 seconds
20:31 (quit) jonrafkind: Ping timeout: 240 seconds
20:33 (quit) penryu: Remote host closed the connection
20:40 (quit) mithos28: Quit: mithos28
20:53 (join) gienah
21:40 (quit) dnolen: Quit: dnolen
21:43 (join) mithos28
21:45 (quit) mithos28: Client Quit
22:07 (join) mithos28
22:09 (quit) mithos28: Client Quit
22:14 (join) realitygrill
22:21 (join) samth_
23:02 (join) ckrailo
23:09 (quit) samth_: Ping timeout: 250 seconds
23:10 (quit) martinhex: Remote host closed the connection
23:27 (join) mithos28