sponge45 changed the topic of #ocaml to: Discussions about the OCaml programming language | http://caml.inria.fr/
swater has quit ["Quat"]
torper has joined #ocaml
<torper> anybody interested in helping me debug? =P
<malc_> torper: you are appending lst in 3 cases and drop it altogether in first two, intended?
johnnowak has joined #ocaml
<torper> yeah
<torper> i imagine i should be using some sort of folding
<torper> but i'm not very good at it yet, and i dont see why this shouldn't work
<malc_> it's not that it wouldn't work
<malc_> it will be slower and more memory hungry than (probably) intended
<torper> ah yes true
<torper> but it also doesn't compile for me: "| Node(left, x, Empty) -> lst @ (helper left Node(Empty, x, parent))
<torper> This expression has type 'a tree but is here used with type 'b list"
<torper> i want helper to return a list, but it seems to think i should have to return a tree?
<malc_> torper: parentise constructor applcation (i.e. (Node(...)))
smimou has quit ["bli"]
torper has left #ocaml []
torper has joined #ocaml
malc_ has quit ["leaving"]
<torper> hmm still seems to complain even if i parenthesise all my constructor applications
<torper> /who malc_
shawn has quit ["This computer has gone to sleep"]
bluestorm has quit ["Konversation terminated!"]
torper has left #ocaml []
johnnowak has quit []
Mr_Awesome has joined #ocaml
jiffyjeff has left #ocaml []
mnemonic has quit ["leaving"]
pango has quit [Remote closed the connection]
pango has joined #ocaml
Mr_Awesome has quit ["...and the Awesome level drops"]
Smerdyakov has quit ["Leaving"]
torper has joined #ocaml
torper has left #ocaml []
Shimei has quit [Read error: 104 (Connection reset by peer)]
Shimei has joined #ocaml
pants1 has quit [Remote closed the connection]
dark_light has quit ["Ex-Chat"]
bluestorm has joined #ocaml
ramkrsna has quit [Read error: 104 (Connection reset by peer)]
ramkrsna has joined #ocaml
shawn has joined #ocaml
_JusSx_ has joined #ocaml
ikaros has joined #ocaml
<flux-> the cuckoo-hash linked at programming.reddit.com seems interesting.. I'm not sure if the ocaml hashing lends itself nicely to it
<flux-> I'm trying to simulate 'secondary hash' with Hashtbl.hash (0, key) but maybe that's a bad choice
<flux-> and maybe I should try to keep the hash length as a prime
love-pingoo has joined #ocaml
love-pingoo has quit [Client Quit]
love-pingoo has joined #ocaml
ikaros has quit [Read error: 110 (Connection timed out)]
ikaros has joined #ocaml
slipstream-- has joined #ocaml
slipstream has quit [Read error: 110 (Connection timed out)]
_JusSx__ has joined #ocaml
diffbavis has quit [Read error: 113 (No route to host)]
_JusSx_ has quit [Read error: 113 (No route to host)]
jlouis_ has joined #ocaml
swater has joined #ocaml
carsten has joined #ocaml
<carsten> Hey, I opened this bug some weeks ago: http://caml.inria.fr/mantis/view.php?id=4143 It has been closed without any comment. I can still reproduce the bug like several other can... A comment would be nice....
<carsten> well, perhaps xleroys comment is the comment I want...
<carsten> ok, forget it
<love-pingoo> carsten: I don't think any caml developper is around
<carsten> yeah, I am jsut testing something
<love-pingoo> we're mostly users
<carsten> love-pingoo: do you by chance have a 64 bit machine?
<love-pingoo> nope
<carsten> that bug only happens on 64 bit
<carsten> and I code on 32
<love-pingoo> what is it ?
<carsten> a crash bug
<love-pingoo> do you use non-standard libraries ? native ints ?
<carsten> no no and no :)
<carsten> comment 3905 seems to be the solution
diffbavis has joined #ocaml
jlouis_ has quit [Remote closed the connection]
<pango> flux-: if the hash functions are uniform, using prime hash table lengths buy you nothing... It's just an old trick around bad hashing functions
carsten has quit ["Konversation terminated!"]
jdev has quit [Read error: 110 (Connection timed out)]
jlouis has joined #ocaml
<flux-> pango, well, how good is ocaml's hash function?
<flux-> I think it is a_new = a_old * alpha + input * beta, which I suppose could be a relatively good hash
<flux-> but not anywhere near md5..
Smerdyakov has joined #ocaml
pango has quit [Remote closed the connection]
pango has joined #ocaml
jdev has joined #ocaml
diffbavis has quit [Read error: 104 (Connection reset by peer)]
pango has quit [Remote closed the connection]
pango has joined #ocaml
diffbavis has joined #ocaml
shawn has quit ["This computer has gone to sleep"]
<flux-> hmh, Hashtbl is embarrassingly faster than my cuckoo hash-attempt - albeit it's with md5, with the ocaml hash it plainly fails because the hash becomes too big because of collisions..
shawn has joined #ocaml
shawn_ has joined #ocaml
<jlouis> flux-, md5 has cryptographic properties you don't need in a hash function
<Smerdyakov> jlouis, isn't it more like "MD5 is _conjectured_ to have cryptographic properties"? :)
<jlouis> Smerdyakov, true
<jlouis> Smerdyakov, But all cryptographical hashes are conjectures
<Smerdyakov> Doesn't change my objection
<jlouis> nope ;)
<flux-> jlouis, yeah, but it was simple to use, because it's in the Digest-module :)
<flux-> and it's easy to extract two different hashes from one value with it..
<flux-> (just extract different parts of the digest=
shawn has quit [Connection timed out]
<jlouis> flux-, I can't read your paragraph about performance... what was fastest -- the cuckoo scheme or the Hashtbl scheme?
<flux-> Hashtbl :)
<flux-> hmph, it takes extra diligence to write hashing functions per the recipes on the net, because of no unsigned 32-bit datatype
<jlouis> You could try MLton, it has a pure 32bit unsigned type
<jlouis> (of course, that feat is possible because of the way polymorphic types are handled in MLton)
<flux-> I suppose it would be _possible_ for ocaml too, but just as inconvenient to use as other non-int-types
<flux-> modern languages just don't seem to support unsigned types
<Smerdyakov> flux-, you use a meaning of "support" that rules out what SML provides?
_JusSx_ has joined #ocaml
<flux-> no, I'm just thinking of java and such
<flux-> (I don't know what the SML support is, but I would guess the same, in principle, as in c)
<flux-> hm, haskell has its Word64-types and such
<flux-> actually now reconsidering my position, maybe java is the only modern statically compiled language without unsigned data types (in addition to ocaml)?-)
_JusSx___ has joined #ocaml
_JusSx__ has quit [Read error: 110 (Connection timed out)]
_JusSx_ has quit [No route to host]
_JusSx_ has joined #ocaml
benny_ has joined #ocaml
johnnowak has joined #ocaml
benny has quit [Read error: 110 (Connection timed out)]
romanoffi has joined #ocaml
jlouis has quit ["Leaving"]
shawn_ has quit ["This computer has gone to sleep"]
love-pingoo has quit [Remote closed the connection]
johnnowak has quit []
love-pingoo has joined #ocaml
jlouis has joined #ocaml
<jlouis> I've played a bit more with OCaml now. It seems rather nice as a language. Very ML-esque ;)
shawn has joined #ocaml
malc_ has joined #ocaml
mnemonic has joined #ocaml
<mnemonic> hi
shawn has quit ["This computer has gone to sleep"]
love-pingoo has quit ["Connection reset by pear"]
shawn has joined #ocaml
romanoffi has quit [Read error: 110 (Connection timed out)]
putpixel has joined #ocaml
putpixel has left #ocaml []
malc_ has quit ["leaving"]