__DL__ changed the topic of #ocaml to: OCaml 3.09.0 available! Archive of Caml Weekly News: http://sardes.inrialpes.fr/~aschmitt/cwn/ | A free book: http://cristal.inria.fr/~remy/cours/appsem/ | Mailing List: http://caml.inria.fr/bin/wilma/caml-list/ | Cookbook: http://pleac.sourceforge.net/
mlh_ has joined #ocaml
shirogane has joined #ocaml
<aegray> still matched it
<aegray> :/
<dylan> weird
<dylan> works for me. :-/
<aegray> hmm
<aegray> | "\"" (( '\\' "\"" | [^'"']) * as str) "\"" {(STRING str)
<aegray> tell me if you see anything blaringly wrong with that
<dylan> I'd use '"'
<dylan> for the first bit.
<aegray> k
Smerdyakov has quit [Read error: 101 (Network is unreachable)]
<aegray> got it by cheating
<dylan> I actually do sublexing on strings now.
<aegray> what does that mean?
Smerdyakov has joined #ocaml
* aegray noob
<dylan> | '"' { dstring [] lexbuf }
<aegray> ah
<dylan> with "and dstring accu = parse"
<aegray> that seems more logical
<dylan> if it helps any, here's a lexer I've been working on: http://svn.hardison.net/bake/trunk/src/lexer.mll
ecc has joined #ocaml
ecc has quit ["Client exiting"]
Bigbang is now known as Bigb[a]ng
ptolomy2 has joined #ocaml
mercurylala has joined #ocaml
mercurylulu has quit [Read error: 104 (Connection reset by peer)]
premchai21 has joined #ocaml
mlh_ has quit [Client Quit]
premchai21 has quit ["Hey! Where'd my controlling terminal go?"]
mercurylulu has joined #ocaml
pango_ has joined #ocaml
mercurylala has quit [Read error: 104 (Connection reset by peer)]
mercurylala has joined #ocaml
mercurylulu has quit [Read error: 104 (Connection reset by peer)]
<KrispyKringle> Anyone here use OMLet?
aegray has quit []
pango has quit [Read error: 110 (Connection timed out)]
Korollary has joined #ocaml
Korollary has left #ocaml []
descender has quit [Read error: 110 (Connection timed out)]
mrpingoo has joined #ocaml
vodka-goo has quit [Read error: 110 (Connection timed out)]
descender has joined #ocaml
khaladan has quit [Connection timed out]
batdog is now known as batdog|gone
batdog|gone is now known as batdog
shirogane has quit [Remote closed the connection]
Snark has joined #ocaml
Snark_Boojum has joined #ocaml
Snark has quit [Read error: 110 (Connection timed out)]
Snark_Boojum is now known as Snark
ramkrsna has joined #ocaml
pango_ has quit ["Client exiting"]
pango has joined #ocaml
vyvepe has joined #ocaml
ppsmimou has joined #ocaml
_fab has joined #ocaml
Bigb[a]ng is now known as Bigbang
ramkrsna has quit [Read error: 110 (Connection timed out)]
_fab has quit [Remote closed the connection]
<revision17_> I finally understand monads
<revision17_> how could I have been so dumb to not realize it
_fab has joined #ocaml
Revision17 has joined #ocaml
pango has quit [Remote closed the connection]
pango has joined #ocaml
revision17_ has quit [Read error: 110 (Connection timed out)]
pango has quit [Remote closed the connection]
pango has joined #ocaml
Skal has joined #ocaml
vezenchio has joined #ocaml
ramkrsna has joined #ocaml
ramkrsna has quit [Read error: 110 (Connection timed out)]
mrpingoo has quit [Read error: 145 (Connection timed out)]
mrpingoo has joined #ocaml
zigong has joined #ocaml
Bigbang is now known as Bigb[a]ng
pingoo has joined #ocaml
<pingoo> hi
<pingoo> I just gave a try at ocamlprof, and I'm a bit disappointed
<pingoo> isn't there a way to get execution speed instead of just number of calls ?
Bigb[a]ng is now known as Bigbang
<pingoo> ah.. if I compile native code I can use gprof, nice
<pingoo> but it runs out of memory ??!
<pingoo> gprof: out of memory allocating 3728792744 bytes after a total of 286720 bytes, whereas the initial program actually runs perfectly
<pingoo> I have 1GB RAM
<pingoo> strange
<mellum> Hmm, profiling works for me
<pingoo> sorry for all that, it was a stupid mistake I did :\
<pingoo> it works, and revealed surprising things, cool
__DL__ has joined #ocaml
mercurylulu has joined #ocaml
mercurylala has quit [Read error: 104 (Connection reset by peer)]
pingoo has quit ["cassos"]
mercurylulu has quit [Read error: 104 (Connection reset by peer)]
mercurylala has joined #ocaml
mercurylala has quit [Read error: 104 (Connection reset by peer)]
mercurylala has joined #ocaml
exa has quit [Read error: 104 (Connection reset by peer)]
exa has joined #ocaml
mercurylala has quit [Read error: 110 (Connection timed out)]
zigong has quit [Read error: 104 (Connection reset by peer)]
mercurylala has joined #ocaml
kmagdsick has joined #ocaml
<kmagdsick> Has there been any talk of changing the Ocaml bytecode to an SSA-based bytecode instead of a stack-based bytecode?
<Smerdyakov> How can a bytecode language be SSA-based?
<kmagdsick> google for llvm
<vincenz> Hello
<kmagdsick> or Michael Franz's SafeTSA bytecode for Java
<kmagdsick> especially read Michael Franz's paper on SafeTSA
* vincenz is stuck on a typing issue
<kmagdsick> hello vincenz
<dylan> vincenz: what's the issue? This one is curious. :)
<vincenz> well
<vincenz> ...
<vincenz> I have elements of type
<vincenz> hmm
<vincenz> how to say this
<dylan> code and pastebin?
<vincenz> yeah
<vincenz> I hope this makes sense
<dylan> You want to have a function whose return type is one of two types?
<vincenz> nono
<vincenz> *makes a second paste*
<vincenz> the type of the list in Appl is dependent on the number of parameters it's funciton get
<dylan> I'm not sure I understand. :(
ramkrsna has joined #ocaml
<dylan> Ah.
<dylan> You can't do that I think.
<vincenz> yeah
<vincenz> but I have to
<dylan> well, you could use a variant.
<dylan> or whatever type foo = Foo of t1 | Bar of t2 ... is called.
<vincenz> yeah I guess
<vincenz> just that I basically bind the arguments of the functioins at the outer level
<vincenz> it should be possible in some way as it sort of reflects how ocaml is made... Just that it seems to me that I'll have to make a different constructor for each different number of argumenst
<vincenz> Fun1, Fun2...
<dylan> if it is a tuple, yes.
<vincenz> alright
ptolomy2 has quit ["Leaving"]
mrpingoo is now known as mpingouin
ramkrsna has quit [Read error: 113 (No route to host)]
ramkrsna_ has joined #ocaml
ppsmimou has quit ["Leaving"]
ramkrsna_ has quit [Remote closed the connection]
smimou has joined #ocaml
pango has quit ["Leaving"]
teem has joined #ocaml
pango has joined #ocaml
<kmagdsick> what's new/cool with 3.09 ?
teem has quit [Read error: 104 (Connection reset by peer)]
shirogane has joined #ocaml
mrsolo has joined #ocaml
Mooninite1 has joined #ocaml
teem has joined #ocaml
teem has quit [Read error: 104 (Connection reset by peer)]
Snark has quit ["Parti"]
<vincenz> anyone use OCamlMakefile?
* dylan waves hand.
<vincenz> How do do subdirs?
<dylan> Not for long though.
<dylan> hmm?
<dylan> That's funny, I just was doing something with subdirs!
<dylan> apparently you can just add, e.g. t/tests.ml to the $SOURCES var.
<vincenz> yeah but...
<mellum> What's the advantage over using a plain Makefile?
<dylan> mellum: all the cruft is hidden in the ocamlmakefile.
<dylan> you just include ../OCamlMakefile in your Makefile and viola, it works.
<mellum> dylan: Hmm. Never really bothered me. Oh well.
<mellum> Probably my project are too simple :)
* dylan notes he is writing a make replacement in ocaml.
<dylan> as soon as bake --write-shellscript works, I'll be able to ditch makefiles. :)
* dylan muttres obscene things about GNU make's patterns in relation to its directed graph of dependencies.
<mellum> dylan: why don't you use one of the 32768 existing make replacements?
<dylan> Because I wanted to write a parser and suchlike
<dylan> and implement lexical scope, first class functions, etc.
<vincenz> dylan: what's wrong with this?
<vincenz> it won't subclean
<dylan> Also, for some reason most other make replacements don't seem to utilize anything from graph theory very well.
<mellum> dylan: what would they need to use besides topological sorting?
<dylan> mellum: well, sticking to the graph I should say.
<dylan> It seems like, for example, gnu make's patterns (%.o: %.c) don't become part of the graph
<dylan> ant is too slow, and... eeew. jam is ugly looking. rake requires ruby...
<mellum> What's this thing KDE is switching to again?
<mellum> something using python
<dylan> Cons
<mellum> so what sucks about cons? ;)
<dylan> it's written in python.
<dylan> also it seems to require more boilerplate than make.
<mellum> *shrug* I like Python. Certainly better than Perl or something.
<dylan> I want something as general purpose as make, but which prevents recursive make and other abominations.
<dylan> really it was an excuse to: learn ocaml, use ocamlyacc et al, and learn more than I ever wanted to know about directed acyclic graphs.
mercurylala has quit ["Leaving"]
Schmurtz has joined #ocaml
<vincenz> makefile really sucks
<vincenz> it's even whitespace sensitive
Bigbang is now known as Bigb[a]ng
<vincenz> bleh
<vincenz> (libraries:%=%/%.cmxa)
<vincenz> it evals the second % as %
<vincenz> not the value
<vincenz> stupid stupid makefile
__DL__ has quit ["Bye Bye"]
mlh_ has joined #ocaml
* dylan is quite impressed at how productive ocaml makes him.
mpingouin has quit []
vyvepe has left #ocaml []
vezenchio has quit ["blupp"]
mercurylala has joined #ocaml
smimou has quit ["bli"]
exa has quit [Remote closed the connection]
<KrispyKringle> mellum: Yes, I'm also a fan of Python. ;)
<KrispyKringle> (Delayed reaction, I know.)
zigong has joined #ocaml