gildor changed the topic of #ocaml to: Discussions about the OCaml programming language | http://caml.inria.fr/ | Inscription for OCaml Meeting 2010 is opened http://wiki.cocan.org/events/europe/ocamlmeetingparis2010
pimmhogeling has quit [Read error: 110 (Connection timed out)]
johnnowak has joined #ocaml
johnnowak has quit [Client Quit]
tmaeda is now known as tmaedaZ
joewilliams has quit [Remote closed the connection]
alexyk has quit []
Drk-Sd` has quit ["dodo"]
djanderson has joined #ocaml
djanderson has quit [Client Quit]
yakischloba has quit ["Leaving."]
alexyk has joined #ocaml
Alpounet has quit [Read error: 113 (No route to host)]
_unK has quit [Remote closed the connection]
lagenar has joined #ocaml
<lagenar> I need to use 16-bit integers. Is there any module for that?
ski_ has joined #ocaml
lagenar has quit ["Saliendo"]
valross has joined #ocaml
ulfdoz has joined #ocaml
fx___ has quit [Read error: 110 (Connection timed out)]
fx_ has joined #ocaml
maskd has quit ["eh"]
yakischloba has joined #ocaml
<thelema> olegfink: the only tiny compilers I know are metaocaml ones.
jeddhaberstro has quit [Client Quit]
joewilliams has joined #ocaml
thrasibule has joined #ocaml
Amorphous has quit [Read error: 110 (Connection timed out)]
Amorphous has joined #ocaml
joewilliams has quit [Remote closed the connection]
ulfdoz has quit [Read error: 110 (Connection timed out)]
Alpounet has joined #ocaml
alexyk has quit []
mehdid has quit [Read error: 104 (Connection reset by peer)]
mehdid has joined #ocaml
Alpounet has quit ["Quitte"]
thrasibule_ has joined #ocaml
thrasibule has quit [Read error: 54 (Connection reset by peer)]
M| has quit [Read error: 60 (Operation timed out)]
ttamttam has joined #ocaml
ttamttam has quit [Client Quit]
yakischloba has quit ["Leaving."]
M| has joined #ocaml
ttamttam has joined #ocaml
ccasin has joined #ocaml
rwmjones_lptp has joined #ocaml
<quelqun_dautre> hi
_zack has joined #ocaml
ccasin has quit [Read error: 110 (Connection timed out)]
pimmhogeling has joined #ocaml
<quelqun_dautre> I have a little question: is there a way to add two floats ? If yes, how ?
<fx_> 1.0 +. 2.0
<quelqun_dautre> thx
<olegfink> thelema: thanks, also there is harrop's minml targetting llvm
rwmjones_lptp has quit [Read error: 60 (Operation timed out)]
<olegfink> well, min-caml is 20 times the size of minml, and I absolutely don't care (for now) about the expressiveness of the language the compiler understand, I need something with simple AST and understandable code generator
rwmjones_lptp has joined #ocaml
rwmjones_lptp is now known as rwmjones
Associat0r has joined #ocaml
Associat0r has quit [Client Quit]
munga has joined #ocaml
<olegfink> if anyone cares, here's a version of miniml modified to work with modern llvm: http://vpaste.net/BDuaq?
fx___ has joined #ocaml
fx_ has quit [Read error: 60 (Operation timed out)]
Yoric has joined #ocaml
caligula_ has joined #ocaml
[df] has joined #ocaml
pimmhogeling has quit ["Ik ga weg"]
pimmhogeling has joined #ocaml
avsm has joined #ocaml
caligula__ has quit [Read error: 110 (Connection timed out)]
<mrvn> fx___: scheme?
<fx___> ?
<mrvn> You can't get much simpler than scheme
ikaros has joined #ocaml
Submarine has joined #ocaml
_andre has joined #ocaml
fx___ has quit [Read error: 110 (Connection timed out)]
fx_ has joined #ocaml
maskd has joined #ocaml
peper has quit ["leaving"]
peper has joined #ocaml
munga has quit ["Ex-Chat"]
avsm has quit ["Leaving."]
ikaros has quit ["Leave the magic to Houdini"]
Yoric has quit []
<olegfink> hm, where should caml_copy_string be defined?
<mrvn> /usr/include/caml/alloc.h:CAMLextern value caml_copy_string (char const *);
<olegfink> well, I meant to ask how should I use it from a binding (when I try to compile some ocaml code with a binding that uses it, I get "undefined reference")
<mrvn> hmm, never had that problem.
ski_ has quit ["Lost terminal"]
<olegfink> okay, here's the code: http://vpaste.net/JH3eQ?
<olegfink> I compile it with $ ocamlc -cc g++ -ccopt -D__STDC_LIMIT_MACROS -ccopt -D__STDC_CONSTANT_MACROS -c modstr.c
<olegfink> and then $ ocamlc -cclib -lstdc++ llvm.cma modstr.o modstr.ml -o modstr which says "undefined reference to `caml_copy_string(char const*)'"
<fx_> put `extern "C"` around includes
<fx_> * around caml includes
<olegfink> thanks, it did it
<olegfink> for some reason I assumed every C header in the universe comes with extern "C" wrapper already
<fx_> not the french ones
<olegfink> yay, now we have string representation of llvm modules!
<olegfink> too bad it can't be done with llvm c api, so it won't make it to the official bindings
Spiwack has joined #ocaml
Submarine has quit [Read error: 110 (Connection timed out)]
ski_ has joined #ocaml
avsm has joined #ocaml
avsm has quit ["Leaving."]
<olegfink> hm, would you recommend going ocamlbuild route to handle all those ocamlc invocation or should I better use a good old Makefile? The code's pretty much standalone, doesn't have to be integrated anywhere.
<julm> ocamlbuild works well out of the box
<olegfink> for llvm?
<julm> and makefile forces you to improve your understanding of the compilation process
<julm> euh
<julm> dunno for llvm
<julm> actually I don't even know what is llvm
<julm> some other dude around will tell you for sure
<olegfink> here's the command line my makefile produces: ocamlc -pp camlp4oof -cclib -lstdc++ -I +camlp4 dynlink.cma camlp4lib.cma llvm.cma llvm_bitwriter.cma llvm_bitreader.cma modstr.o minml.ml -o minml
<olegfink> can I get the same without myocamlbuild.ml?
<julm> does one of the .cma have the -cclib -lstdc++ in its embedded flags?
<[df]> I want to experiment with a multi-module project in the top-level - do I need to build each module to a .cmo and then #load it?
<olegfink> probably not, because it doesn't get picked if I omit it from the ocamlc invocation
<julm> olegfink: then I don't know, wait for the answer of somebody familiar with ocamlbuild
zhijie has quit ["Leaving."]
zhijie has joined #ocaml
<julm> [df]: you could create a module in the toplevel as in a file with the [module] keyword
<fx_> It it is the llvm_*.cma modules that need -lstdc++ - then it would be better to embed this option inside that cma
<[df]> julm: you mean put them all in the same module?
<fx_> and then ocambuild without plugin would do
<[df]> I want to leave them in different modules for when I build to an executable
derdon has joined #ocaml
<fx_> otherwise you need a plugin to drive it
<julm> [df]: then use several [module] keywords
<julm> module A = struct end;;
<julm> module B = struct end;;
<julm> whatever you want
<[df]> I'm not sure I follow
<[df]> I'd put that at the top of each file?
<julm> what do you want?
pimmhogeling has quit [Read error: 110 (Connection timed out)]
<[df]> I've got this project which currently builds to an executable
<[df]> but I want to easily test it from the toplevel
<fx_> [df], you can load a .cma
<[df]> what I'm doing currently is using a separate Makefile rule to build the .cmo files, and then a script to load them all into the toplevel
<julm> yep using a .cma sounds well
<julm> ocamlc -a -o out.cma *.cmo
<julm> then: ocaml out.cma
<[df]> great, thanks
pimmhogeling has joined #ocaml
<olegfink> ok, http://vpaste.net/8kLiP? is probably terser than whatever ocamlbuild can do
<julm> oh vpaste :)
<julm> I think ocamlbuild provides some easy way to get the -l
<julm> -get+pass
<olegfink> what's with vpaste? :-)
<julm> it's vim :)
<julm> and it's small
<julm> and free software
<julm> as in freedom
<olegfink> I can't export a module that imports everything from another module?
<olegfink> so to speak module B = {A with foo = something}
<Spiwack> include A
<mrvn> A with foo = something is something for functors I believe. Not for module inclusion.
<Spiwack> It's for functor types.
<Spiwack> But I think he was aluding to the record with-notation.
<julm> olegfink: there is a patch out there to do that, but the masters have never merged it upstream
<Spiwack> no but seriously. "include A" works.
<[df]> is there a way to run arbitrary code from within the debugger?
<Spiwack> Or I didn't understand the question.
<julm> but include A does not overrides modules or types
<mrvn> Spiwack: I think you did.
<julm> -s
<olegfink> yes, include seems to do what I want
<mrvn> julm: which you can only do with functors
<julm> functors, yep
<olegfink> yay!
<olegfink> it wouldn't be a wise idea to have a file a.ml which starts with "include A", right?
<mrvn> would that even work?
<julm> depends what you'll do with your resulting A
<Spiwack> Well, in general having two files named the same is a good way to have ocaml *really* confused.
<mrvn> and yourself
<julm> I think packed .cmo could help solve conflicts
_zack has quit ["Leaving."]
ikaros has joined #ocaml
<olegfink> there's an Llvm module in /usr/lib/ocaml, and I want to locally redefine it (adding another function)
<olegfink> yes, that wasn't a bright idea
<mrvn> module MyLlvm = struct include Llvm let foo x = x end
<julm> you could also do: module Sup = struct module Llvm = Llvm end module Llvm = struct let foo x = Sup.Llvm.foo x end
<olegfink> yes, that works, naming it LLvm doesn't.
<olegfink> *Llvm
<olegfink> okay, I think I'll just rename it to Myllvm or something like that
<mrvn> LLvm should work too
<olegfink> can anyone remind me of the usual .depend techinque for ocaml?
tmaedaZ is now known as tmaeda
<julm> ocamldep -I +toto *.ml > .depend ?
<olegfink> well, the Makefile part
<julm> include .depend ?
<olegfink> I remember -include .depend, but how to ensure it's always generated first?
<julm> make depend
<olegfink> only explicit?
<julm> which calls ocamldep
<julm> well, you do it by hand after a change yup
<julm> you could also put it in the all: rules
<julm> dunno what you want
gim has quit [Remote closed the connection]
<julm> you could put the *.ml as dependences of the depend rules so that it will only be executed when the .ml are more recent than the .depend
<julm> bof
<olegfink> ...but let's try :-)
<julm> .depend: $(wildcard *.ml)\n\tocamldep $(OCAMLDEPFLAGS) $^ > $@
<olegfink> yes, the "scary" refers to the proposed %.d scheme
<julm> that it is an implicit rule
<julm> % is a joker
<julm> you could access it with $*
<olegfink> (with %.d make regenerates them automatically via makefile rewriting)
<olegfink> hm, interesting, it .generates .depends by itself
<olegfink> though I didn't tell it to: http://vpaste.net/XZgxw?
Submarine has joined #ocaml
<julm> olegfink: the include rule needs the file
<julm> perhaps that's why
alexyk has joined #ocaml
<olegfink> hm, gnu make doesn't support arrays as variables so I can't say something like OCAMLFLAGS(parser.ml)=-pp camlp4oof?
_zack has joined #ocaml
<olegfink> and then %: %.ml\n\tocamlc $(OCAMLFLAGS($<)) ...
<julm> no that I'm aware of
<julm> not*
<julm> perhaps you could use the arrays of your shell
<julm> setting SHELL=zsh for instance
<olegfink> I think gnu autoconf uses something like OCAMLFLAGS_parser.ml=... but that's realling getting pretty ugly
<julm> indeed
<Spiwack> Then again, if you're looking a some clean programming language for your compilation script.
<julm> ocaml
<Spiwack> You might start being interested in ocamlbuild.
<julm> Spiwack: +1
<olegfink> yeah, let's make a deal: someone rewrites my current Makefile into myocamlbuild.ml, I start using it. :-P
<julm> huhu
<flux> olegfink, and said rewriting doesn't simply invoke make?-)
<olegfink> flux, Makefile should be gone by the time :-)
fx_ has quit [Read error: 110 (Connection timed out)]
<julm> http://git.ocamlcore.org/cgi-bin/gitweb.cgi < you could be inspired by the myocamlbuild.ml of these projects
* thelema needs to finish his ocamlbuild-plugins-directory patch
fx_ has joined #ocaml
<Camarade_Tux> \o/
<julm> oh a ctux
<julm> ctux, for you: 15:39:29 olegfink : yeah, let's make a deal: someone rewrites my current Makefile into myocamlbuild.ml, I start using it. :-P
<julm> I know you know
<julm> ctux, http://vpaste.net/XZgxw < the makefile
alexyk has quit []
<Camarade_Tux> julm: use cam<tab> ;-)
<Camarade_Tux> julm: hmmm, not today ;-)
<julm> too loong
<julm> flemmard
Submarine has quit [Read error: 110 (Connection timed out)]
Submarine has joined #ocaml
<olegfink> hrrr
joewilliams has joined #ocaml
<olegfink> http://vpaste.net/EetXF? -- syntax.ml http://vpaste.net/UEyTg? -- parser.ml
<olegfink> ocamlc -pp camlp4oof -I +camlp4 -c parser.ml slvm.cma syntax.cmo -o parser.cmo
<olegfink> File "parser.ml", line 10, characters 8-19:
<olegfink> Error: Unbound constructor If
<olegfink> how come?
<olegfink> if I do #load "syntax.cmo" in toplevel, Syntax.If is there just fine
ulfdoz has joined #ocaml
<olegfink> ah, cool, module name conflict
<julm> olegfink: thereis a Syntax in preCast
<julm> erf, grilled
<olegfink> hrm, isn't .cma enough? it seems it also wants .cmi for the module
<julm> .cmi is mandatory yes
valross has quit ["Ex-Chat"]
Yoric has joined #ocaml
Submarine has quit [Read error: 110 (Connection timed out)]
avsm has joined #ocaml
Submarine has joined #ocaml
<olegfink> hm, I'm doing something wrong
<olegfink> despite that I have .depend, I have to list the .cmo's to link with in the "right" order, otherwise ocamlc doesn't find them
<julm> yep
<julm> -> ocamlbuild
<julm> alternatively you could use tsort on a slightly modified .depend to deduce the topological sequence
ikaros has quit ["Leave the magic to Houdini"]
<olegfink> okay, I'll just wait till Camarade_Tux writes myocamlbuild.ml for me :-)
<julm> mouarf
<julm> tsort is not hard
<thelema> olegfink: check out the Ocamlviz module starting at L170 in this: http://codepad.org/a7dekC7R
<thelema> all it does is add libocamlviz.cm?a to the compile and link when there's an ocamlviz tag, but it might be a good place to start.
<thelema> I have no idea if the ocamlviz_auto works - I wrote it but never used it
<olegfink> aha
alexyk has joined #ocaml
yakischloba has joined #ocaml
Drk-Sd` has joined #ocaml
alexyk has quit []
Spiwack has quit ["Leaving"]
alexyk has joined #ocaml
alexyk has quit [Client Quit]
alexyk has joined #ocaml
_unK has joined #ocaml
hcarty has quit [lindbohm.freenode.net irc.freenode.net]
WuJiang has quit [lindbohm.freenode.net irc.freenode.net]
rbancroft has quit [lindbohm.freenode.net irc.freenode.net]
msteele has quit [lindbohm.freenode.net irc.freenode.net]
ertai has quit [lindbohm.freenode.net irc.freenode.net]
peddie has quit [lindbohm.freenode.net irc.freenode.net]
fremo_ has quit [lindbohm.freenode.net irc.freenode.net]
Leonidas has quit [lindbohm.freenode.net irc.freenode.net]
mbishop has quit [lindbohm.freenode.net irc.freenode.net]
infoe has quit [lindbohm.freenode.net irc.freenode.net]
hcarty_ has joined #ocaml
msteele has joined #ocaml
WuJiang_ has joined #ocaml
infoe has joined #ocaml
Leonidas has joined #ocaml
rbancroft has joined #ocaml
peddie has joined #ocaml
mbishop has joined #ocaml
ertai_ has joined #ocaml
fremo_ has joined #ocaml
alexyk has quit []
<olegfink> julm: sed 's/://' .depend | tsort | grep -v cmx | tac seems to work
alexyk has joined #ocaml
<julm> hum
<julm> aaah tac is cool
<olegfink> but, but, it gives me all the modules, e.g. it has several entry points
<julm> olegfink: I was using something more complicated: http://vpaste.net/bLnEC
<Camarade_Tux> olegfink: not sure what the problem is but do you know 'uniq'?
<olegfink> yes, but uniq doesn't know which entry point I need
<olegfink> hm
<olegfink> I should always use CAMLparam in a function, right?
alexyk has quit []
snowbeard has quit [Read error: 110 (Connection timed out)]
<julm> if you use the parameters yep
avsm has quit ["Leaving."]
<olegfink> so http://vpaste.net/Jk8aG? is wrong?
<julm> yep
<julm> CAMLparam/CAMLreturn
<julm> I think
<julm> especially when there is a alloc_small in the middle
<olegfink> that'd explain why it corrupts things.
<julm> maybe :)
<olegfink> (it corrupts ModuleID)
<julm> that said, maybe llvm brings/deletes something special that makes this CAML* useless
tmaeda is now known as tmaedaZ
<olegfink> it seems strange, LLVMModuleRef is opaque, ocaml has no idea about what it is
<Camarade_Tux> doubtful and probably not worth it, and it should be easy to test it anyway ;-)
<olegfink> then somewhere in it there is a pointer to ocaml string
<olegfink> and then ocaml gc's the string...
<olegfink> well, llvm takes about an hour to build
<julm> ctux: ^ and on your phantom?
<Camarade_Tux> olegfink: I have a machine that should build it pretty quickly but I don't have much time, so if you can give me what I need to download and what I need to do, I can do it ;-)
<Camarade_Tux> (no need to go into too many details though ;-) )
<Camarade_Tux> btw, I'll put up a machine meant for compiling and testing across various OS and architectures (qemu though, but quad-core) at least for the mingw-w64. And for ocaml too unless everyone tell me noone will use it: if you want to save me _some_ time, tell it now ;-)
<Camarade_Tux> btw, this machine compiles gcc in 5 minutes :-)
<julm> go go go ctux !
<olegfink> Camarade_Tux: which distro is it running?
<julm> slack I guess
<Camarade_Tux> olegfink: I use slackware64 but I'll put various virtual machines (I'm currently even downloading a ppc32 distrib to see how fast [or slow] it is)
alexyk has joined #ocaml
ttamttam has quit ["Leaving."]
<olegfink> very very strange
<olegfink> can't reproduce the corruption by hand
<olegfink> Camarade_Tux: you can use http://repos.archlinux.org/wsvn/community/llvm/repos/community-i686/PKGBUILD as a reference probably
<Camarade_Tux> olegfink: is clang a requisite for you?
<olegfink> nope, ignore it
derdon has quit [Read error: 104 (Connection reset by peer)]
derdon has joined #ocaml
alexyk has quit []
alexyk has joined #ocaml
_zack has quit ["Leaving."]
ccasin has joined #ocaml
<Camarade_Tux> 1109.29user 71.61system 7:24.74elapsed 265%CPU
<Camarade_Tux> it was making it with only 'make -j1' at first so I had to cancel it ;-)
<Camarade_Tux> olegfink: so, what to test?
<thelema> odd... how can this raise [Not_found]? [ if q.length = 0 then raise Empty;
<julm> thelema: exception Empty = Not_found ?
<thelema> Fatal error: exception Not_found
<thelema> Raised at file "queue.ml", line 87, characters 29-34
<thelema> this is being raised by my command-line argument code in the middle of my program's execution!
<thelema> I guess the backtrace could be wrong
<julm> yep
* thelema tries bytecode...
<julm> Empty != Not_found
<olegfink> Camarade_Tux: http://vpaste.net/DZTqg?
<olegfink> prints garbage for me
<thelema> ah, much better. a Not_found being raised at a totally different location
<Camarade_Tux> thelema: would it be possible you have a mix of modules? like ocaml picking the bad one? or different versions?
pad` has joined #ocaml
<thelema> It's probably related to me updating my local batQueue module...
<thelema> although I would expect that to raise an error
<thelema> maybe the signature didn't change.
<Camarade_Tux> olegfink:
<Camarade_Tux> ; ModuleID = 'p�'
<olegfink> cool, now the course of action is probably to add CAMLprim/CAMLreturn to llvm_create_module in llvm/bindings/ocaml/llvm/llvm_ocaml.c
<Camarade_Tux> (the second character enclosed in single quotes is unreadable)
<Camarade_Tux> okay, trying
<olegfink> (ModuleID should be what I gave it, e.g. "name")
<olegfink> s/e.g./i.e./
<Camarade_Tux> ok
ccasin has quit [Remote closed the connection]
<olegfink> maybe adding printf("%s\n", String_val(ModuleID)) in llvm_create_module would help
<Camarade_Tux> who wrote the bindings?
<Camarade_Tux> I'm not a pro of C bridges but this one looks really weird: it's not returning something of type "value"
<olegfink> not me!
<Camarade_Tux> ^^
<olegfink> you know, some of its arguments aren't of type value either
<julm> |* Note that these functions intentionally take liberties with the CAMLparamX *|
<julm> |* macros, since most of the parameters are not GC heap objects. *|
<Camarade_Tux> well spotted
<Camarade_Tux> doesn't cost much to try
<olegfink> by the way I tried the other way 'round
<olegfink> wrote a wrapper around C++ function to set the module id
<olegfink> even without CAML* and Gc.compact() between setting and printing everything looks fine
hyperboreean has quit [Read error: 110 (Connection timed out)]
ccasin has joined #ocaml
<Camarade_Tux> julm: regarding the comment you pasted, that works for params but not for return values, right?
alexyk has quit []
<julm> euh
<Camarade_Tux> casting to value didn't help, I get the same output
<Camarade_Tux> unfortunately I really can't spend more time on that today and I have to go =/
<Camarade_Tux> (will be free tomorrow)
alexyk has joined #ocaml
* julm reads caml/memory.h
<olegfink> Camarade_Tux: thanks, have fun
ccasin has quit [Remote closed the connection]
Yoric has quit []
alexyk has quit []
demitar has joined #ocaml
demitar has quit [Read error: 104 (Connection reset by peer)]
alexyk has joined #ocaml
_andre has quit ["*puff*"]
alexyk has quit []
Submarine has quit [Read error: 110 (Connection timed out)]
alexyk has joined #ocaml
ttamttam has joined #ocaml
itewsh has joined #ocaml
gilsonm has joined #ocaml
ttamttam has quit ["Leaving."]
alexyk has quit []
ygrek has joined #ocaml
kotrin has joined #ocaml
kotrin has left #ocaml []
ulfdoz has quit [Read error: 110 (Connection timed out)]
itewsh has quit [Client Quit]
otswim has joined #ocaml
<otswim> hello, has someone even used the implementation of fibheap for ocaml? i implemented dijkstra, but sometimes i get "Fatal error: exception FibHeap.Make(Ord).Empty" when doing fibheap_extract_min_data heap whereas fibheap_size heap outputs a lot more than 0; has anyone ever had this problem?
alexyk has joined #ocaml
CcSsNET has quit [K-lined]
jcaose has joined #ocaml
anders^^ has quit [Read error: 60 (Operation timed out)]
derdon has quit []
ygrek has quit [Remote closed the connection]
<otswim> i'm beginning to think that's because i'm not compiling fiblib correctly, what is the good way to do that? to 'make' the src and then?
Submarine has joined #ocaml
<otswim> what should i do with the cma and cmxa? just use them within ocamlopt?
jcaose has quit [Read error: 110 (Connection timed out)]
<Camarade_Tux> cma with ocamlc and cmxa with ocamlopt
<Camarade_Tux> maybe more a problem in the fib-heap implementation but I don't know, I never used it
<olegfink> bleh, ocaml is a beatiful language, but a terrible toolkit
ski_ has quit ["Lost terminal"]
travisbrady has joined #ocaml
<olegfink> is now too late to just redo the whole system? :-)
<olegfink> (before godi and batteries get more widespread)
<olegfink> Camarade_Tux: you have automated gtk+ wrapper generator, right?
<thelema> olegfink: batteries isn't sufficient - an improved ocamlbuild might be
<Camarade_Tux> olegfink: not complete and not perfect though
<Camarade_Tux> right now, it does webkit-gtk really well but I know others which don't completely work
<olegfink> well, with all my love to unix, I believe such an adanced language must have a different compilation model, probably more like java's (though it's very weird at certain points)
<travisbrady> is anyone aware of an sql parser for ocaml? I have an svn repo with a few thousand sql queries in it and I'm hoping to find duplicates and near duplicates
<olegfink> Camarade_Tux: I'm just fed up with current llvm ocaml bindings and feel advanterous enough to try something from scratch
<travisbrady> I googled but just found this YACAMLQT which is a C# project
<olegfink> or maybe I should just screw llvm
ikaros has joined #ocaml
<thelema> travisbrady: I don't know of one - you've checked the ocaml hump?
<thelema> olegfink: It seems everyone is doing llvm now
<travisbrady> thelema: i have, but i always have difficulty navigating around there so maybe i missed it
<travisbrady> thelema: btw, nice work on batteries, i spent some time this weekend investigating it
<thelema> travisbrady: fair enough. I don't know of one, but expect a large subset easy to parse
<olegfink> it's a bit like webkit -- a millions times betters than gecko but still a huuge pile of c..p
<thelema> travisbrady: thanks.
<travisbrady> would it be reasonable to attempt to build one using BatParserCo?
<olegfink> doesn't pgocaml come with some camlp4-based parser?
<travisbrady> ahh, i'll have a look at that. thanks olegfink
<Camarade_Tux> olegfink: there are two things: 1- ocaml-gir relies on some "glib" "conventions" (rather, conventions followed by glib-based libraries) but it should be possible to overcome that problem, 2- gobject-introspection is a horrible POS, I wouldn't mind dumping it now even though I've solved the problem i had with it
hyperboreean has joined #ocaml
<Camarade_Tux> just needs a good C (C++ ?) parser
<otswim> Camarade_Tux: i can't do open FibHeap if i compile with ocamlopt fiblib.cmxa etc.
<olegfink> nice joke, Camarade_Tux :-)
<Camarade_Tux> ;-)
<Camarade_Tux> olegfink: and remember that you can't bind "int*" without annotations ;-)
<otswim> http://pastebin.com/f7176a5b8: maybe do i need some other files that just fiblib.cmxa in order to compile? for instance some of the fibheap.*?
<olegfink> just why people _start_ writing stuff in C++? (like llvm)
<Camarade_Tux> so that you have a harder time writing binding to it ;-)
fremo_ has quit [lindbohm.freenode.net irc.freenode.net]
ertai_ has quit [lindbohm.freenode.net irc.freenode.net]
TaXules has quit [lindbohm.freenode.net irc.freenode.net]
companion_cube has quit [lindbohm.freenode.net irc.freenode.net]
noj has quit [lindbohm.freenode.net irc.freenode.net]
ertai has joined #ocaml
TaXules has joined #ocaml
fremo_ has joined #ocaml
companion_cube has joined #ocaml
alexyk has quit []
Submarine has quit [Read error: 113 (No route to host)]
<otswim> has someone ever heard of the exception: FibHeap.Option.No_value; i'm getting it but this exception is document nowhere, i've looked the doc of fiblib and didn't find this exception in it (it has been raised by fibheap_decrease_key) and google has 0 result for it
ikaros has quit ["Leave the magic to Houdini"]
palomer has joined #ocaml
tmaedaZ is now known as tmaeda