| --- Log opened Wed Sep 19 00:00:01 2007 |
| 00:09 | | gnolam [lenin@Nightstar-10613.8.5.253.static.se.wasadata.net] has quit [Quit: Goddamn plague.] |
| 00:13 | | You're now known as TheWatcher[T-2] |
| 00:15 | | You're now known as TheWatcher[zZzZ] |
| 00:32 | | Netsplit DeepThought.NY.US.Nightstar.Net <-> Troika.TX.US.Nightstar.Net quits: Vornicus, MinceR, @Thaqui, ReivZzz, jerith, @Chalain |
| 00:32 | | MinceR [~mincer@Nightstar-21862.nik.bmf.hu] has joined #code |
| 00:32 | | mode/#code [+o Thaqui] by ChanServ |
| 00:32 | | Netsplit over, joins: Thaqui |
| 00:32 | | Netsplit over, joins: ReivZzz |
| 00:32 | | Vornicus [~vorn@64.252.103.ns-11691] has joined #code |
| 00:33 | | Vornicus is now known as NSGuest-1126 |
| 00:35 | | NSGuest-1126 is now known as Vornicus |
| 00:37 | | Chalain [~chalain@Nightstar-4184.res.logixcom.net] has joined #code |
| 00:37 | | jerith [~jerith@Nightstar-21563.slipgate.za.net] has joined #code |
| 00:38 | | jerith is now known as NSGuest-1129 |
| 00:47 | | * McMartin plays with OCaml |
| 00:47 | <@McMartin> | Hilarity with abstract types. |
| 00:47 | <@McMartin> | I'm writing a "factory method" that converts a primitive type into what a certain abstract type secretly is. |
| 00:47 | <@McMartin> | The function to do this is: |
| 00:47 | <@McMartin> | let convert x = x;; |
| 00:48 | <@McMartin> | And that is how you do type-casting in a language with type inference. |
| 00:49 | < Vornicus> | ...okay. |
| 00:49 | <@McMartin> | (The actual cast is in the stated signature elsewhere.) |
| 00:57 | | ReivZzz is now known as ReivClass |
| 01:19 | < MinceR> | gn |
| 01:44 | <@McMartin> | val test : (ByteStream.StringStream.t -> 'a) -> ('a -> string) -> string -> unit |
| 01:45 | <@McMartin> | I'm glad I didn't have to declare that. |
| 01:56 | <@ToxicFrog> | o.O |
| 01:56 | <@ToxicFrog> | I should play with ML someday, but then I see stuff like that and it scares me. |
| 01:57 | <@ToxicFrog> | What does that actually mean? I mean, the first two chunks look like they declare a function that maps functions with a given signature to functions with a different signature, but then there's two more -> and I get lost. |
| 02:29 | | Thaqui [~Thaqui@Nightstar-26933.jetstream.xtra.co.nz] has left #code [Leaving] |
| 02:36 | <@McMartin> | TF: The part you're missing is that ML functions only usually bother with one argument, because you can express multiple arguments with chained invocations. |
| 02:37 | <@McMartin> | And, in fact, use this to do partial evaluation. |
| 02:37 | <@McMartin> | That is a function that takes two functions and a string and returns nothing. |
| 02:38 | <@McMartin> | The first function takes a user defined type and returns... something. The second is a function that converts that something into a string. The third is an annotation. |
| 02:38 | | * McMartin is reimplementing Python's struct module in OCaml so that he could perhaps in the future actually use OCaml for something. |
| 03:19 | <@McMartin> | Woohoo! Byte-exactness! |
| 03:19 | <@McMartin> | Even for floating point! |
| 03:19 | <@McMartin> | No support for unsigned quad words on the OCaml side, but even Python doesn't guarantee quadword support of any kind, so. |
| 03:25 | | GeekSoldier|Bed [~Rob@Nightstar-5191.pools.arcor-ip.net] has quit [Ping Timeout] |
| 03:27 | <@ToxicFrog> | Floating point is the persistent ball of ;.; in my Lua struct implementation. |
| 03:29 | <@McMartin> | So, yeah |
| 03:29 | | * McMartin works on building a makefile to demonstrate everything. |
| 03:30 | <@McMartin> | This is, of course, comically imperative code. |
| 03:33 | <@ToxicFrog> | Heh. |
| 03:33 | <@ToxicFrog> | The thing that always messes me up in pure-functional is the lack of state. |
| 03:33 | <@ToxicFrog> | Or, rather, the lack of explicit state. |
| 03:33 | <@McMartin> | Nnnnormally that's not a problem for me. |
| 03:33 | <@McMartin> | However, when one is writing I/O code and one is not using Haskell, one is mostly out of luck. |
| 03:33 | | * ToxicFrog is all about the mutable data structures |
| 03:39 | | * ToxicFrog grabs Java and shakes it until the ability to mutate the class tree falls out |
| 03:40 | <@McMartin> | Ew |
| 03:40 | <@McMartin> | Morphing is evil and wrong |
| 03:47 | <@ToxicFrog> | Well, Doing It Right involves subclassing String. |
| 03:47 | <@ToxicFrog> | ...and then rewriting the compiler. |
| 03:47 | <@McMartin> | ... or just using StringBuilders everywhere in your code? |
| 03:47 | <@ToxicFrog> | StringBuilders? |
| 03:48 | <@McMartin> | The non-constant version of String? |
| 03:48 | <@ToxicFrog> | All I want is to add one method to String. Is that so hard? |
| 03:48 | <@McMartin> | Which method would that be? |
| 03:48 | <@ToxicFrog> | (hint: yes, if you want the strings generated by string literals and every goddamn library function that returns strings to be able to use it) |
| 03:48 | <@ToxicFrog> | reverse(). |
| 03:49 | <@ToxicFrog> | And, yeah, I can always do class ReversableString extends String, but that doesn't help me write someRandomLibraryFunction().toString().reverse() |
| 03:49 | <@McMartin> | ... and writing a static method somewhere else that takes a String and returns a new one isn't good enough for this? |
| 03:49 | <@ToxicFrog> | Well, yes, I can do that, but that's ugly. |
| 03:50 | <@McMartin> | I see nothing wrong with TFUtil.reverse(someRandomLibraryFunction().toString()) -- or at least, nothing worse than what you bring upon yourself from writing in Java to begin with. |
| 03:50 | <@McMartin> | If you want pretty code write in Scheme~ |
| 03:50 | <@ToxicFrog> | Sadly, my prof is a dick. |
| 03:51 | <@McMartin> | http://www.stanford.edu/~mcmartin/misc/camlstruct.tgz |
| 03:52 | <@McMartin> | The function "test" with the scary type signature is in camlread.ml. |
| 03:52 | <@ToxicFrog> | (and I find that foo.reverse(bar.baz) is less intuitively obvious than bar.baz.reverse) |
| 03:53 | <@ToxicFrog> | Anyways. I don't actually need this, it would just be nice to have. |
| 03:56 | <@McMartin> | (I note, in passing, that OCaml has horrific syntax and the Wrong Holes in its type system, but has the Good and Portable compilers.) |
| 03:58 | | * McMartin is going to use this so that there's an OCaml-based library for SVAF. |
| 03:58 | < Vornicus> | what are its holes? |
| 03:58 | <@McMartin> | The comparators and the hash function are magic. |
| 03:58 | <@McMartin> | And can fall down as a result on user-defined types. |
| 03:59 | <@McMartin> | However, + and - are not magic and so if you want to do math on floating point numbers you must use +. and -. instead. |
| 03:59 | < Vornicus> | ow. |
| 03:59 | <@McMartin> | Yeah. |
| 03:59 | <@McMartin> | Also, int is "As much of an integer as I can fit into my tagged memory model." |
| 03:59 | <@McMartin> | So on most systems that means an int is 31 bits. |
| 04:00 | <@McMartin> | My little struct module handles the necessary conversions under the hood for that. |
| 04:00 | <@McMartin> | (There are other types for 32- and 64-bit No I Mean It integers.) |
| 04:00 | < Vornicus> | Ruby also has that, but does transparent conversion to longs. |
| 04:00 | < Vornicus> | er, bigintegers. |
| 04:00 | <@McMartin> | ML is Brutally Statically Typed. |
| 04:01 | <@McMartin> | But it can get away with it due to polymorphism of madness and type inference. |
| 04:01 | <@McMartin> | The upshot of that is that it basically never needs to do a classical method dispatch. |
| 04:25 | | GeekSoldier|Bed [~Rob@Nightstar-5191.pools.arcor-ip.net] has joined #code |
| 04:26 | | GeekSoldier|Bed is now known as GeekSoldier|work |
| 04:39 | | * McMartin starts commenting his code. |
| 04:39 | <@McMartin> | "This implementation is imperative like whoa, but what the heck, so are channels." |
| 06:28 | | Thaqui [~Thaqui@Nightstar-26933.jetstream.xtra.co.nz] has joined #code |
| 06:28 | | mode/#code [+o Thaqui] by ChanServ |
| 06:30 | | NSGuest-1129 [~jerith@Nightstar-21563.slipgate.za.net] has quit [Quit: leaving] |
| 06:30 | | jerith [~jerith@Nightstar-21563.slipgate.za.net] has joined #code |
| 07:02 | | Vornicus is now known as Vornicus-Latens |
| 07:17 | | GeekSoldier|work [~Rob@Nightstar-5191.pools.arcor-ip.net] has quit [Ping Timeout] |
| 08:31 | | Vornotron [~vorn@69.0.81.ns-26780] has joined #code |
| 08:32 | | Vornicus-Latens [~vorn@ServicesOp.Nightstar.Net] has quit [Ping Timeout] |
| 08:36 | | GeekSoldier|work [~Rob@Nightstar-5191.pools.arcor-ip.net] has joined #code |
| 08:40 | | ReivClass is now known as ReivOut |
| 09:01 | | You're now known as TheWatcher |
| 09:31 | | ToxicFrog [~ToxicFrog@Admin.Nightstar.Net] has quit [Operation timed out] |
| 09:32 | | ToxicFrog [~ToxicFrog@Admin.Nightstar.Net] has joined #code |
| 09:32 | | mode/#code [+o ToxicFrog] by ChanServ |
| 11:09 | | Chalcedon [~Chalcedon@Nightstar-2472.ue.woosh.co.nz] has quit [Quit: Gone] |
| 12:45 | | Thaqui [~Thaqui@Nightstar-26933.jetstream.xtra.co.nz] has left #code [Leaving] |
| 13:22 | | ReivOut is now known as Reiver |
| 13:44 | | gnolam [lenin@Nightstar-10613.8.5.253.static.se.wasadata.net] has joined #Code |
| 13:44 | | mode/#code [+o gnolam] by ChanServ |
| 14:48 | | Pi [~sysop@Nightstar-6875.hsd1.wa.comcast.net] has joined #code |
| 14:48 | | mode/#code [+o Pi] by ChanServ |
| 14:55 | < MinceR> | r4wr |
| 15:12 | | GeekSoldier|work is now known as GeekSoldier |
| 15:14 | | Reiver is now known as ReivZzz |
| 15:37 | | GeekSoldier [~Rob@Nightstar-5191.pools.arcor-ip.net] has quit [Ping Timeout] |
| 16:21 | | GeekSoldier [~Rob@Nightstar-5191.pools.arcor-ip.net] has joined #code |
| 16:32 | | Vornotron is now known as Vornicus |
| 18:42 | | GeekSoldier [~Rob@Nightstar-5191.pools.arcor-ip.net] has quit [Ping Timeout] |
| 18:43 | | GeekSoldier [~Rob@Nightstar-4842.pools.arcor-ip.net] has joined #code |
| 18:48 | | gnolam [lenin@Nightstar-10613.8.5.253.static.se.wasadata.net] has quit [Ping Timeout] |
| 19:12 | < EvilDarkLord> | Any apache gurus around? |
| 19:34 | | gnolam [lenin@Nightstar-10613.8.5.253.static.se.wasadata.net] has joined #Code |
| 19:34 | | mode/#code [+o gnolam] by ChanServ |
| 19:35 | | Forj [~Forj@Nightstar-2472.ue.woosh.co.nz] has joined #code |
| 19:35 | | mode/#code [+o Forj] by ChanServ |
| 19:54 | | Forj [~Forj@Nightstar-2472.ue.woosh.co.nz] has quit [Quit: Gone] |
| 20:07 | | GeekSoldier is now known as GeekSoldier|bed |
| 21:14 | | Chalcedon [~Chalcedon@Nightstar-2472.ue.woosh.co.nz] has joined #code |
| 21:14 | | mode/#code [+o Chalcedon] by ChanServ |
| 23:01 | | Thaqui [~Thaqui@Nightstar-26933.jetstream.xtra.co.nz] has joined #code |
| 23:01 | | mode/#code [+o Thaqui] by ChanServ |
| 23:41 | | You're now known as TheWatcher[t-2] |
| 23:44 | | You're now known as TheWatcher[zZzZ] |
| 23:49 | < MinceR> | gn |
| --- Log closed Thu Sep 20 00:00:08 2007 |