00:05 (join) dented42
00:06 (nick) gridaphobe -> muppetbot
00:09 asumu: Shambles_: I just use the printf function. And yeah, that's what I ended up doing.
00:10 (quit) rbarraud: Read error: Connection reset by peer
00:22 (join) jrslepak
00:34 (join) kreol[Ukr]
00:44 (join) veer
00:51 mithos28: Is anyone here a TR user? I want to know whether they think (let: ((x : (U (Vectorof Number) String) (vector 1 2 3))) x) should typecheck? This is not does it typecheck but should it.
01:02 (quit) veer: Read error: Connection reset by peer
01:03 (join) veer
01:07 veer: mithos28: it type checks here in 5.2.1
01:07 (quit) karswell: Remote host closed the connection
01:07 mithos28: veer: that is not the question. The question is do you expect it to type check?
01:08 veer: mithos28: yes
01:08 mithos28: why? I am interested in your reasoning
01:08 mithos28: A change I am making breaks this and I want to see if I agree in trying to see if it should be supported
01:10 veer: let returns x and x is of type T , so it will type check
01:10 veer: i.e let expr
01:10 mithos28: The interesting part is (vector 1 2 3) of type (U (Vectorof Number) String)
01:11 jrslepak: should it not be?
01:12 mithos28: jrslepak: It depends
01:12 mithos28: (vector 1 2 3) without any context generalizes to (Vectorof Integer)
01:12 mithos28: which is not (U (Vectorof Number) String)
01:12 veer: Integer is subtype of Number
01:13 (quit) jonathansizz: Ping timeout: 245 seconds
01:13 (quit) getpwnam: Ping timeout: 245 seconds
01:13 mithos28: but vectors are mutable and thus invariant not covariant
01:13 mithos28: If you replace vector with list there is no question that this should typecheck
01:15 veer: I don't know what invariant and covariant means sorry, but when you mutate x , x can only have values of types that are subtype of number as far as i know
01:16 veer: I mean number or string
01:17 mithos28: (Vectorof Integer) is not a subtype of (Vectorof Number)
01:17 mithos28: because if I had a vector of integers and treated it as a vector of numbers I could set some index to a non-integer
01:17 (join) karswell
01:18 mithos28: which would cause problems to something that thought every index was an integer
01:18 (quit) kreol[Ukr]: Ping timeout: 244 seconds
01:18 veer: when you set some index to non-integer where integer is expected than type checker will produce error
01:20 mithos28: right, but if (vectorof integer) was a subtype of (vectorof Number) you could get away with this. Which is why it is not.
01:20 (quit) Kaylin: Read error: Connection reset by peer
01:21 veer: thing is number is not subtype of integer and if index a[i] say and i is of type integer than if you try to set i to a non integer it will cause type error
01:21 mithos28: rudybot: (define v (vector 1 2 3))
01:21 rudybot: mithos28: your sandbox is ready
01:21 rudybot: mithos28: Done.
01:22 mithos28: rudybot: (define v2 v)
01:22 rudybot: mithos28: Done.
01:22 mithos28: (rudybot (vector-set! v2 0 4.2)
01:22 mithos28: rudybot: (vector-set! v2 0 4.2)
01:22 rudybot: mithos28: Done.
01:22 mithos28: rudybot (vector-ref v 0)
01:23 mithos28: rudybot: (vector-ref v 0)
01:23 rudybot: mithos28: ; Value: 4.2
01:23 mithos28: If v was type (Vectorof Integer) and v2 was type (Vectorof Number), the only possible error line is the definition of v2
01:24 mithos28: that is why (Vectorof Integer) is not a subtype of (Vectorof Number)
01:26 veer: let me put a concrete example , i have forgotten how to annotate types , so I'll come back in few mins :)
01:29 veer: mithos28: you are right
01:30 veer: so why is it  so ?
01:30 mithos28: Ok but note that you can still annotate (vector 1 2 3) as (Vectorof Number) because 1 2 3 can be considered numbers
01:31 mithos28: If the context is expecting a vector of numbers (vector 1 2 3) will say that it returns a vector of numbers
01:32 mithos28: but if the context doesn't specify then it generalizes to integers
01:33 mithos28: The real question is in the case where there is a union type do you expect TR to go into the union type and find the one vector type
01:33 mithos28: and what should it do if there are multiple vector types
01:34 mithos28: currently there is a lot of special casing, but I don't think it actually helps any real programs
01:34 mithos28: and it dosen't work if you use vector-prime where (define vector-prime vector)
01:39 veer: my question is Number type eqv to (U Integer Float .... Number) ?
01:40 mithos28: There is no Number in the union. But yes Number is a union of more precise numeric types
01:41 veer: so (Vectorof Number) eqv to (Vectorof (U Integer Float ...)) ?
01:41 mithos28: Yep
01:41 veer: so why (define v2 v1) produces real error ?
01:42 veer: v2: (Vectorof Number)
01:42 mithos28: let a < b mean that a is a subtype of b
01:42 veer: ok
01:42 mithos28: while Integer < Number, (Vector Integer) < (Vector Number) is not true
01:43 veer: I see
01:43 mithos28: Thats part of the definition of the Vector type
01:43 mithos28: if it was true, then you could get errors
01:43 veer: ok
01:44 mithos28: also if it was true you would be programming Java
01:44 mithos28: they got it wrong
01:46 veer: ok thanks , any example of that can cause error if it was true?  I can't seem to come up with any ?
01:47 mithos28: The example I gave earlier
01:47 mithos28: you treat a vector of integers as a vector of numbers
01:47 mithos28: you then set on of the entries to a number
01:47 mithos28: you then look at through the vector of integers value and look at that entry
01:48 veer: ok ok I got it now :)
01:48 mithos28: it is a floating point number and not an integer
01:48 veer: so this is because of mutation
01:48 mithos28: sorta
01:49 mithos28: it is really because of the argument appearing on both sides of the function arrow
01:49 mithos28: it appears on the left in set! and on the right in ref
01:51 veer: thanks mithos28 , I was really surprised why it did not work , just checked it woks with list thanks
01:54 (join) mithos28_
01:54 (nick) mithos28_ -> mithos28
01:58 (quit) karswell: Remote host closed the connection
02:01 (join) mithos28_
02:01 (nick) mithos28_ -> mithos28
02:02 (quit) jacius: Quit: Leaving
02:06 (join) mithos28_
02:06 (quit) mithos28_: Remote host closed the connection
02:08 (join) karswell
02:08 (join) mithos28
02:19 (join) adu
02:32 (quit) adu: Quit: adu
02:37 (quit) mceier: Quit: leaving
02:46 (quit) mithos28: Remote host closed the connection
02:46 (join) mithos28
02:47 (quit) dsantiago: Ping timeout: 252 seconds
02:52 (join) dsantiago
02:52 veer: One more thing I notice is that if I  let v1 be a list of integers [1 2 3] and v2 be a vector of anything  say number, when I do this (vector-set! v2 (first v1) 3) it does not type check , but (vector-ref v2 (first v1)) does
02:53 mithos28: can you paste code?
02:54 veer: ok in a sec
02:56 veer: https://gist.github.com/2940271
02:57 mithos28: rudybot: init typed/racket
02:57 rudybot: mithos28: your typed/racket sandbox is ready
02:58 mithos28: rudybot: (define v2 (vector 1 2 3))
02:58 rudybot: mithos28: Done.
02:58 mithos28: rudybot: (:print-type v2)
02:58 rudybot: mithos28: ; stdout: "(Vector (U 0 1 Byte-Larger-Than-One Positive-Index-Not-Byte Positive-Fixnum-Not-Index Negative-Fixnum Positive-Integer-Not-Fixnum Negative-Integer-Not-Fixnum) (U 0 1 Byte-Larger-Than-One Positive-Index-Not-Byte Positive-Fixnum-Not-Index Negative-Fixnum Positive-Integer-Not-Fixnum Negative-Integer-Not-Fixnum) (U 0 1 Byte-Larger-Than-One Positive-Index-Not-Byte Positive-Fixnum-Not-Index Negative-Fixnum Pos
02:59 mithos28: rudybot: (vector-set! v2 (+ 1 1) 4)
02:59 rudybot: mithos28: error: #<undefined>:1:16: Type Checker: expected statically known index for heterogeneous vector, but got Positive-Index in: (#%app + (quote 1) (quote 1))
02:59 mithos28: Are you getting that error or something else?
02:59 veer: yes similar
02:59 (join) mithos28_
02:59 (quit) mithos28: Read error: Connection reset by peer
02:59 (nick) mithos28_ -> mithos28
03:00 (quit) jonrafkind: Ping timeout: 248 seconds
03:00 (join) neilv
03:00 mithos28: so v2 is not a (Vectorof Integer) in this case it is (Vector Integer Integer Integer)
03:00 mithos28: so it is statically of length 3 so typedracket likes to add extra checks
03:01 veer: oh ok
03:01 mithos28: rudybot: (define v-set vector-set!)
03:01 rudybot: mithos28: Done.
03:01 mithos28: rudybot: (v-set v2 (+ 1 1) 4)
03:01 rudybot: mithos28: Done.
03:01 mithos28: rudybot: v2
03:01 rudybot: mithos28: ; Value: #(1 2 4)
03:01 rudybot: mithos28: ; stdout: "- : (Vector Integer Integer Integer)\n"
03:02 mithos28: but the checks are awful hacks, that can be avoided by not using the vector-set! binding
03:02 mithos28: But thanks for bringing that up, I look to fix it
03:03 mithos28: the issue is that it just gives up in the non static index case, when it should see if it could do in to all of the arguments
03:08 (join) mithos28
03:11 veer: allright thanks  , it is specific to vector-set! ,  vector-ref seems to work both with static index and dynamic
03:12 mithos28: Its because it is easier, it is a union type in vector-ref where it is the interesection in vector-set!
03:17 (join) jesyspa
03:20 (nick) eMBee -> em-bee
03:29 (join) mithos28_
03:29 (quit) mithos28: Read error: Connection reset by peer
03:29 (nick) mithos28_ -> mithos28
03:30 (join) mithos28_
03:30 (nick) mithos28_ -> mithos28
03:46 (join) cataska
04:00 (join) mithos28_
04:01 (quit) Shambles_: Quit: Leaving.
04:04 (nick) mithos28_ -> mithos28
04:06 (join) mceier
04:24 (quit) veer: Quit: Leaving
04:47 (join) bitonic
04:49 (quit) karswell: Remote host closed the connection
05:00 (join) karswell
05:10 (join) MayDaniel
05:16 (quit) MayDaniel: Read error: Connection reset by peer
05:22 (join) antithesis
05:41 (quit) antithesis: Quit: antithesis
05:44 (quit) jesyspa: Quit: leaving
05:53 (join) antithesis
05:58 (join) Shambles_
06:12 (quit) sizz_: Quit: No Ping reply in 180 seconds.
06:12 (join) sizz
06:16 (quit) antithesis: Quit: antithesis
06:20 neilv: i have enough working in charterm that i'm going to document and release it now. http://i.imgur.com/p1qwX.png
06:26 (quit) bitonic: Quit: WeeChat 0.3.7
06:28 (join) bitonic
07:12 Shambles_: neilv: Now everybody can make roguelikes in Racket. ;)
07:14 neilv: yes, i just documented that fact 10 minutes ago
07:19 (join) lamefun3
07:19 (join) masm
07:20 (part) lamefun3
07:25 (join) jonathansizz
07:25 (join) getpwnam
07:37 (join) RacketCommitBot
07:37 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/YsKDDw
07:37 RacketCommitBot: [racket/master] add a possible explanation for indy - Robby Findler
07:37 (part) RacketCommitBot
07:40 neilv: i once wrote a roguelike in pascal or c
07:40 neilv: 3 of them, actually
07:42 (quit) karswell: Remote host closed the connection
07:52 (join) karswell
08:05 (join) veer
08:19 (quit) Shviller: Ping timeout: 248 seconds
08:19 (join) Shviller
08:39 (quit) veer: Ping timeout: 240 seconds
08:41 neilv: why is there no "nonnegative?"
08:54 Shambles_: (define (nonnegative? x) (or (zero? x) (positive? x)))
09:06 neilv: i wanted to use it in a contract in documentation
09:08 neilv: i don't know how people can stand to document things like scribble
09:08 neilv: i'm suffering just documenting this little library i wrote over the course of a day or so
09:22 (join) MayDaniel
09:31 (join) veer
09:56 (join) baconwithfries
09:58 baconwithfries: heya! Can someone help me with Intermediate lang in DrRacket? I got this error when I try to run a programm : module: identifier already imported from a different source
09:58 baconwithfries: Settings are on standard
10:03 (quit) baconwithfries: Quit: Page closed
10:09 (quit) ChanServ: shutting down
10:25 (join) ChanServ
10:35 (join) mithos28
10:35 (join) jeapostrophe
10:35 (quit) jeapostrophe: Changing host
10:35 (join) jeapostrophe
10:44 (join) kreol[Ukr]
10:49 asumu: neilv: (>=/c 0) is useful for that.
10:56 neilv: http://www.neilvandyke.org/racket-charterm/
10:58 asumu: neilv: Nice!
10:58 neilv: thanks
10:58 neilv: i'm wondering whether that demo command line in the introduction will actually work for anyone
11:00 asumu: Works for me
11:00 neilv: excellent. thanks
11:04 (quit) jrslepak: Quit: This computer has gone to sleep
11:04 jamessan: how well does it handle key chords?  can it tell the difference between <Tab> and <C-i> (in a properly configured terminal)?
11:05 neilv: it knows a lot of them, including things like alt-a, ctrl-f1, shift-f1
11:05 neilv: it can't distinguish tab and ctrl-i, however
11:05 neilv: at least not with the current decoder dfa
11:05 neilv: and not with my rxvt
11:05 jamessan: libtermkey might be useful to look at
11:07 neilv: thanks, i've made a note to look at that
11:09 (join) jacius
11:12 (nick) _p4bl0 -> p4bl0
11:20 (join) jrslepak
11:30 (quit) jrslepak: Quit: This computer has gone to sleep
11:30 (join) RacketCommitBot
11:30 RacketCommitBot: [racket] plt pushed 2 new commits to master: http://git.io/pbH03w
11:30 RacketCommitBot: [racket/master] Don't export internal names for define & lambda. - Sam Tobin-Hochstadt
11:30 RacketCommitBot: [racket/master] Handle submodules in Typed Racket. - Sam Tobin-Hochstadt
11:30 (part) RacketCommitBot
11:47 (join) jonrafkind
11:47 (quit) jonrafkind: Changing host
11:47 (join) jonrafkind
11:52 (join) kreol
11:53 (nick) kreol -> Guest35125
11:54 (quit) kreol[Ukr]: Ping timeout: 260 seconds
12:12 chandler: neilv: what is the shell out to stty used for?
12:21 (quit) MayDaniel: Read error: Connection reset by peer
12:22 (quit) veer: Ping timeout: 246 seconds
12:23 mithos28: Does anyone know why if I raise an error with source location it doesn't get printed, but raise-syntax-error does print it? Is there something I need to do besides attach the list of srclocs using the structure property.
12:25 neilv: chandler: to set raw mode and turn off echo
12:31 (join) Kaylin
12:37 mithos28: Ugh, this is because the source location in a raise-syntax-error is part of the error message
12:38 mithos28: why does the default handler not print the source location
12:39 jonrafkind: are you saying (error) vs (raise-syntax-error) ?
12:40 jonrafkind: or (raise-syntax-error) with an extra argument used for source location?
12:40 mithos28: I am trying to use error to raise an error with source location so that it is printed nicely and aid in debugging
12:41 mithos28: raise-syntax-error was displaying nice sourcelocation information but I couldn't figure out why error was not
12:41 mithos28: and the answer is that the source location information is part of the message for raise-syntax-error
12:41 jonrafkind: so its a runtime error right, not a syntax error?
12:42 mithos28: yes
12:43 mithos28: rudybot: (exn-message (with-handlers ((exn:fail? identity)) (raise-syntax-error 'foo "an error" #'here)))
12:43 rudybot: mithos28: ; Value: "#<undefined>:1:89: foo: an error in: here"
12:43 rudybot: mithos28: ; stdout: "- : String\n"
12:43 jonrafkind: well i mean whats the point, the source location will be from the implementation of the function instead of the call to the function
12:43 jonrafkind: unless you're debugging the implementation i guess
12:44 mithos28: It is in the output of a macro (match to be precise) so it will have the right source location
12:45 jonrafkind: oh so you want the source location to point at the match expression?
12:45 jonrafkind: or possibly some sub-expression inside the match?
12:45 mithos28: the root problem is that the default error-display-handler doesn't use the source location information on the exception in its output
12:45 mithos28: I have already attached the source location to the exception, but I am no longer getting useful output from it
12:46 (quit) karswell: Remote host closed the connection
12:46 mithos28: I made the original change a year ago (according to github) but something has changed in the meantime
12:46 jonrafkind: oh
12:47 mithos28: it is really annoying when it says that no clause matched for some input, but I have no idea what match form it is talking about
12:49 jonrafkind: rofl. story of my life. since all i do is write macros i get no source location for any errors whatsoever
12:50 mithos28: cannot you use syntax/loc?
12:50 jonrafkind: thats for a syntax error
12:50 jonrafkind: try doing (define-syntax (foo stx) (syntax->datum #f)) or something
12:50 jonrafkind: all you get is 'syntax->datum: expected syntax'
12:50 mithos28: so you need contracts at the syntax-level
12:51 mithos28: are you working on a layer below syntax/parse?
12:51 jonrafkind: contracts dont usually have source location information in them either
12:51 (join) netrino
12:51 jonrafkind: ill try an experiment..
12:51 mithos28: they at least have the module usually
12:51 jonrafkind: lol
12:52 mithos28: unless you are using them incorrectly because reexporting contracted values is impossible to do correctly and easily
12:52 jonrafkind: then i should probably put each function into its own module
12:52 (quit) Nisstyre: Quit: Leaving
12:52 mithos28: use define/cond-contract from TR
12:53 mithos28: it uses a variable at phase 1 to determine whether or not the function is contracted
12:53 (quit) dented42: Quit: Computer has gone to sleep.
12:53 mithos28: this allows you to turn them on for debugging but remove them in 'production'
12:54 jonrafkind: where is that, i just grepped for it
12:54 mithos28: collects/typed-racket/utils/utils.rkt
12:55 (quit) Kaylin: Quit: Leaving.
12:55 jonrafkind: oh i was looking in typed/
12:55 jonrafkind: whats the TR directory these days? typed-racket?
12:55 mithos28: oh and it is currently broken, the import of racket/contract/base needs to be switched to racket/contract
12:56 (join) Nisstyre
12:56 mithos28: I have no idea what the directory situation is
12:56 (join) karswell
12:57 jonrafkind: ok if i make a submodule that provides a contracted `foo' and use it in the outer module incorrectly, the error does not mention the submodule
12:58 mithos28: What if you do it in two seperate modules?
12:58 jonrafkind: isn't it already in two separate modules?
12:58 jonrafkind: you mean separate files?
12:58 mithos28: top level modules
12:58 mithos28: I haven't played around with submodules
12:58 asumu: jonrafkind: try `define/contract`
12:59 jonrafkind: oh, i was using provide/contract
12:59 asumu: well, I meant as an alternative to submodules.
12:59 jonrafkind: same thing
12:59 jonrafkind: the point of using submodules was to see if the contract error would mention the submodule
13:01 mithos28: what is the blaming line?
13:01 jonrafkind: without using errortrace it just says 'x.rkt [running body]'
13:01 mithos28: https://gist.github.com/2941947
13:02 mithos28: That is what I get which blames b correctly
13:03 jonrafkind: oh oops, i was getting a different error because i was calling the function with the wrong number of values
13:03 jonrafkind: but the contract error only gets invoked if you use the correct arity but the wrong types
13:03 jonrafkind: why doesn't the contract error fire for mismatched arity too?
13:04 mithos28: Thats probably because the error happens before the contract wrapper
13:04 jonrafkind: do you think it can be changed?
13:04 mithos28: otherwise all contracted procedures would have to accept any number of arguments in the wrapper
13:05 jonrafkind: maybe there is some contract form that can work that way?
13:05 mithos28: Theoretically yes, I'm not sure what the runtime cost would be
13:05 jonrafkind: so not everyone has to use it
13:06 mithos28: acutally no you cannot do it in a contract
13:06 mithos28: it would change procedure-arity
13:06 jonrafkind: oh wait wait, the contract error gives you the source location for the definition of the contracted thing, not the calling thing
13:06 mithos28: right, but gives you the module of the error
13:06 jonrafkind: bleh
13:08 jonrafkind: wait what? it doesnt give me the module of the error
13:08 jonrafkind: it gives me the module of the contracted function
13:08 mithos28: can you paste your output
13:08 jonrafkind: https://gist.github.com/2941967
13:08 jonrafkind: caling function is x.rkt
13:09 mithos28: what are the files?
13:09 jonrafkind: x.rkt and foo.rkt
13:09 mithos28: and contents?
13:10 jonrafkind: updated
13:11 mithos28: Yeah thats right
13:11 mithos28: the contract boundary is the define not the provide
13:12 mithos28: so you are protecting the definition of the function from the use of the function outside
13:12 jonrafkind: as opposed to what
13:12 mithos28: you want to be protecting it from misuse outside the module
13:12 jonrafkind: how should i change my definition of foo or whatever
13:13 mithos28: you need to use contract-out
13:13 jonrafkind: or provide/contract, right?
13:13 mithos28: yeah
13:13 jonrafkind: i thought i was providing the contracted value
13:13 mithos28: you are
13:13 jonrafkind: and doesnt provide/contract also provide a contracted value?
13:13 mithos28: they are two different contracted values
13:14 mithos28: when you contract a value there are two parties that are involved in the contract
13:14 jonrafkind: ok now it says 'blaming x'
13:14 mithos28: with define/contract it is the definition context and the body of the function
13:14 jonrafkind: well this isn't a great state of affairs but i guess its better than nothing
13:15 mithos28: with provide/contract it is the importing module and exporting module
13:15 jonrafkind: ok so i can use define/contract and provide/contract at the same time, thats fine
13:16 mithos28: yeah you will pay the overhead twice, but protect against two different kinds of errors
13:16 mithos28: there really needs to be a provide form recontract-out
13:17 jonrafkind: which would do what
13:17 mithos28: change the blame party to the importing module
13:18 mithos28: but with the same contract
13:18 jonrafkind: oh as the define/contract
13:18 jonrafkind: yea
13:19 mithos28: its requires some rework of the system because the current binding of a contracted value is not inspectable to determine enough information to do that
13:31 (quit) bitonic: Quit: WeeChat 0.3.7
13:32 mithos28: Added my own custom error-display-handler and got a very nice error message
13:32 mithos28: <collects>/typed-racket/typecheck/tc-metafunctions.rkt:74.2:
13:32 (join) bitonic
13:38 (join) yoklov
13:38 Shambles_: neilv: I had to go AFK a long time.  I don't really mind documenting things.  I used to keep up some articles on a wiki.  Felt nice keeping other people from having to suffer through screwing up a bunch to figure out what I'd already figured out.  Must just be a taste thing.
13:43 Shambles_: Also, this isn't a big deal, but is there some way to attach icons to things in the list and tree GUI widgets?  I can't seem to find a way.
13:55 (quit) bitonic: Remote host closed the connection
14:06 (join) adu
14:06 (join) MayDaniel
14:12 (quit) mithos28: Remote host closed the connection
14:13 (join) mithos28
14:21 (join) dented42
14:21 (join) snearch
14:29 (quit) dented42: Quit: Computer has gone to sleep.
14:31 (quit) Myk267: Quit: Myk267
14:45 (quit) adu: Quit: adu
15:00 (quit) snearch: Quit: Verlassend
15:16 (join) banseljaj
15:18 (quit) neilv: Quit: Leaving
15:25 (join) jrslepak
15:33 banseljaj: Hey guys.
15:34 banseljaj: I have a question about DrRacket.
15:34 banseljaj: I am going through the htdp book, and I can't load teachpacks.
15:34 banseljaj: DrRacket v 5.1.3
15:36 Shambles_: banseljaj: I'm not sure if this is the issue, but the latest is 5.2.1
15:37 banseljaj: I was using the one available in ubuntu repositroy
15:37 banseljaj: I will try witht the newer one
15:51 (join) dented42
15:53 mithos28: are there contracts for the dictionary datatypes?
15:54 (quit) jeapostrophe: Ping timeout: 240 seconds
16:06 (nick) banseljaj -> imami|afk
16:11 Shambles_: mithos28: Maybe I don't understand the question, but can't you just attach the contracts to any variable or function, whatever the type?
16:12 mithos28: Shambles_: You cannot write hash/c without having access to the internals of hash tables are implemented
16:12 mithos28: I cannot find such a contract dict/c for dictionaries
16:25 (join) mithos28_
16:29 (quit) mithos28: Ping timeout: 240 seconds
16:29 (nick) mithos28_ -> mithos28
16:30 Shambles_: Guess I have some reading to do.  I thought dictionaries and hash tables were the same thing.
16:33 (join) Myk267
16:35 mithos28: Shambles_: Dictionaries are a generalization of hash tables. Other things that are dictionaries are alists and vectors along with free-id-tables and such
16:35 Shambles_: mithos28: I see that now.  So what are you having trouble with?  You're trying to implement something for Typed Racket?
16:36 mithos28: No trying to debug it, and realizing that dictionary contracts would help
16:39 Shambles_: mithos28: So part of the implementation lacks a contract?
16:39 mithos28: Yes.
16:40 mithos28: I am doing a fairly invasive change to the internals of it so causing lots of errors where I have updated one side of the api but not the other
16:42 Shambles_: mithos28: Was there a bug with the way they worked before, or is there something new you're adding?
16:43 mithos28: I'm trying to improve its inference system. It is not technically a bug, but it feels like it as a user of TR.
16:46 Shambles_: mithos28: I see.
16:50 (quit) MayDaniel: Read error: Connection reset by peer
16:54 (quit) dented42: Quit: Computer has gone to sleep.
16:59 (quit) getpwnam: Ping timeout: 252 seconds
16:59 (quit) jonathansizz: Ping timeout: 252 seconds
17:01 (join) dented42
17:01 (join) MayDaniel
17:05 (join) zkl
17:06 (part) zkl
17:56 (quit) MayDaniel: Read error: Connection reset by peer
18:08 (join) Kaylin
18:32 (join) tfb
18:56 (join) RacketCommitBot
18:56 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/-SLnNQ
18:56 RacketCommitBot: [racket/master] Disable debugging printing. - Vincent St-Amour
18:56 (part) RacketCommitBot
18:59 (nick) eli -> elibarzilay
19:10 (quit) mceier: Quit: leaving
19:13 (quit) tfb: Quit: sleeping
19:26 (quit) netrino: Quit: Ave!
19:45 (quit) sizz: Quit: No Ping reply in 180 seconds.
19:47 (quit) peeeep: Quit: Konversation terminated!
19:48 (join) peeeep
19:48 (join) sizz
19:50 (quit) antono: *.net *.split
20:24 (quit) dented42: Quit: Computer has gone to sleep.
20:24 (quit) masm: Quit: Leaving.
20:25 (quit) stchang: Read error: Operation timed out
20:28 (join) stchang
21:06 (join) hemmie
21:09 hemmie: hi all I have a quick question: I'm trying to get started with drracket, but when I start it up, none of the buttons at the top of the window are showing. They do seem to work normally, but visually it's just a grey bar.. Was wondering if this is at all a familiar problem with drracket, or whether I should be looking for the problem on my end
21:10 hemmie: all menus and buttons on other programs seem to be working normally, and the problem persists after rebooting and reinstalling the package.
21:10 hemmie: This is on ubuntu 12.04 23bit
21:19 (quit) karswell: Remote host closed the connection
21:20 (join) RacketCommitBot
21:20 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/xWc-Lg
21:20 RacketCommitBot: [racket/master] scribble: adjust `def...' box to show a background label in HTML - Matthew Flatt
21:20 (part) RacketCommitBot
21:29 (join) karswell
21:38 (quit) hemmie: Ping timeout: 246 seconds
21:58 (join) hemmie
22:05 (part) hemmie: "WeeChat 0.3.5"
22:11 (quit) Kaylin: Read error: Connection reset by peer
22:29 asumu: mithos28: there is prop:dict/contract but it's a weird interface.
22:29 asumu: And may warrant a re-design in the long run.
22:46 (quit) jacius: Quit: Leaving
23:05 (join) Kaylin
23:24 (join) getpwnam
23:24 (join) jonathansizz
23:55 asumu: The new labels for documenation forms is nice.
23:55 asumu: Though the color slightly hard to see.
23:56 asumu: (it helps a lot on some of the framework docs where it's hard to tell what is a mixin, method, etc.)