tag:www.nabble.com,2006:forum-14000Nabble - Caml2008-10-15T12:39:49ZCaml is a general-purpose programming language, designed with program safety and reliability in mind. It is very expressive, yet easy to learn and use. Caml supports functional, imperative, and object-oriented programming styles. It has been developed and distributed by INRIA, France's national research institute for computer science, since 1985. Caml home is <a href="http://caml.inria.fr/" target="_top" rel="nofollow">here</a>.tag:www.nabble.com,2006:post-20000748Re: OCaml version 3.11.0+beta12008-10-15T12:39:49Z2008-10-15T12:39:49ZAndres Varon<br>On Oct 15, 2008, at 12:03 PM, Xavier Leroy wrote:
<br><div class='shrinkable-quote'><br>>>> Native dynlink used to work on Mac OS X < 10.5 (x86 only). The new
<br>>>> linker in 10.5 does not support linking shared libraries with non-
<br>>>> PIC
<br>>>> code. It is still possible to use the old linker, called ld_classic,
<br>>>> but some libraries (like X11) does not work, so this has been
<br>>>> disabled
<br>>>> in the configure script.
<br>>>>
<br>>>> The clean solution to make natdynlink work on recent Mac OS X
<br>>>> systems
<br>>>> (beside convincing Apple to support the old behavior of their linker
<br>>>> in their new implementation) is to change OCaml's x86 backend so
<br>>>> that
<br>>>> it produces only PIC code (this has been done for the AMD64 port). I
<br>>>> don't think there is currently any plan to work on that.
<br>>>
<br>>> Ouch, this makes it almost a dead end for us. I can offer some time
<br>>> to
<br>>> help in this effort, working in the port, or providing feedback. The
<br>>> native dynlink and toplevel are, at least to me, the killer
<br>>> features in
<br>>> 3.11, but adding another hole for Mac OS X intel (in addition to not
<br>>> supporting x86_64) does not seem like the best choice for an
<br>>> increasingly popular architecture.
<br>>
<br>> Well, we'd very much like to support native dynlink on OS X 10.5,
<br>> but Apple is not helping in the least by crippling their linker
<br>> compared with the one in 10.4. If anyone from Apple is on this list,
<br>> feel free to contact us at <a href="http://www.nabble.com/user/SendEmail.jtp?type=post&post=20000748&i=0" target="_top" rel="nofollow">caml-devel@...</a> for more
<br>> information on this regression.
</div><br>I'm glad to report that I just tried with XCode 3.1.1 and the linker
<br>did not complain (as it did with XCode 3.0). I wrote a little test and
<br>the native Dynlink worked with that version of XCode (using 10.5.5,
<br>Intel -- I had to drop an -I flag in the linker call from the OCaml
<br>compiler when I was trying it).
<br><br>Andres
<br><br>>
<br>>
<br>> - Xavier Leroy
<br>>
<br><br>_______________________________________________
<br>Caml-list mailing list. Subscription management:
<br><a href="http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list" target="_top" rel="nofollow">http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list</a><br>Archives: <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>Beginner's list: <a href="http://groups.yahoo.com/group/ocaml_beginners" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners</a><br>Bug reports: <a href="http://caml.inria.fr/bin/caml-bugs" target="_top" rel="nofollow">http://caml.inria.fr/bin/caml-bugs</a><br><p>From forum: <a href="http://www.nabble.com/Caml-Discuss2-f14016.html" embed="fixTarget[14016]" target="_top" >Caml Discuss2</a></p>tag:www.nabble.com,2006:post-20000406Re: "ocaml_beginners"::[] What's the names of that construct: let s .... s2008-10-15T12:21:33Z2008-10-15T12:21:33ZFlorent Monnier> Anyway, here's the One True Way to read all lines from a file into
<br>> memory:
<br>>
<br>> <a href="http://camltastic.blogspot.com/2008/09/tip-read-all-lines-from-file-most.ht" target="_top" rel="nofollow">http://camltastic.blogspot.com/2008/09/tip-read-all-lines-from-file-most.ht</a><br>>ml
<br><br>if the file is not a text file, the function input_line is maybe not the
<br>recomanded one to use.
<br>Also sometimes a generated text file is not right segmented by newline
<br>characters.
<br><br>Without further informations about the task, we can not tell more about using
<br>this or that method.
<br><br><br>------------------------------------
<br><br>Archives up to December 31, 2007 are also downloadable at <a href="http://www.connettivo.net/cntprojects/ocaml_beginners/" target="_top" rel="nofollow">http://www.connettivo.net/cntprojects/ocaml_beginners/</a><br>The archives of the very official ocaml list (the seniors' one) can be found at <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>Attachments are banned and you're asked to be polite, avoid flames etc.Yahoo! Groups Links
<br><br><*> To visit your group on the web, go to:
<br> <a href="http://groups.yahoo.com/group/ocaml_beginners/" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners/</a><br><br><*> Your email settings:
<br> Individual Email | Traditional
<br><br><*> To change settings online go to:
<br> <a href="http://groups.yahoo.com/group/ocaml_beginners/join" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners/join</a><br> (Yahoo! ID required)
<br><br><*> To change settings via email:
<br> mailto:<a href="http://www.nabble.com/user/SendEmail.jtp?type=post&post=20000406&i=0" target="_top" rel="nofollow">ocaml_beginners-digest@...</a>
<br> mailto:<a href="http://www.nabble.com/user/SendEmail.jtp?type=post&post=20000406&i=1" target="_top" rel="nofollow">ocaml_beginners-fullfeatured@...</a>
<br><br><*> To unsubscribe from this group, send an email to:
<br> <a href="http://www.nabble.com/user/SendEmail.jtp?type=post&post=20000406&i=2" target="_top" rel="nofollow">ocaml_beginners-unsubscribe@...</a>
<br><br><*> Your use of Yahoo! Groups is subject to:
<br> <a href="http://docs.yahoo.com/info/terms/" target="_top" rel="nofollow">http://docs.yahoo.com/info/terms/</a><br><br><p>From forum: <a href="http://www.nabble.com/Ocaml-Beginner-f14002.html" embed="fixTarget[14002]" target="_top" >Ocaml Beginner</a></p>tag:www.nabble.com,2006:post-20000307Re: "ocaml_beginners"::[] Beginner wonders about constructs: let s ...s, with ...2008-10-15T12:16:13Z2008-10-15T12:16:13ZRichard Jones-4On Wed, Oct 15, 2008 at 11:11:25AM -0000, ben.aurel wrote:
<br>> A. what does that construct let s = ... s mean and how is it called?
<br>> Is it something like function where the second 's' is the closing bracket?
<br><br>Maybe this helps?
<br><br><a href="http://ocaml-tutorial.org/the_structure_of_ocaml_programs" target="_top" rel="nofollow">http://ocaml-tutorial.org/the_structure_of_ocaml_programs</a><br><br>Rich.
<br><br>--
<br>Richard Jones
<br>Red Hat
<br><p>From forum: <a href="http://www.nabble.com/Ocaml-Beginner-f14002.html" embed="fixTarget[14002]" target="_top" >Ocaml Beginner</a></p>tag:www.nabble.com,2006:post-20000272Re: "ocaml_beginners"::[] What's the names of that construct: let s .... s2008-10-15T12:14:10Z2008-10-15T12:14:10ZRichard Jones-4<br>I didn't get the full original message - it seems that yahoogroups is
<br>stripping attachments or something.
<br><br>Anyway, here's the One True Way to read all lines from a file into
<br>memory:
<br><br> <a href="http://camltastic.blogspot.com/2008/09/tip-read-all-lines-from-file-most.html" target="_top" rel="nofollow">http://camltastic.blogspot.com/2008/09/tip-read-all-lines-from-file-most.html</a><br><br>However if you want to read a file and do something with it, such as
<br>copy it somewhere else, then it's better not to read it all into
<br>memory. Maybe you should do:
<br><br> run "cp %s %s" (Filename.quote orig) (Filename.quote output)
<br><br>using the trick here:
<br><br> <a href="http://camltastic.blogspot.com/2008/09/writing-printf-like-functions.html" target="_top" rel="nofollow">http://camltastic.blogspot.com/2008/09/writing-printf-like-functions.html</a><br><br>Rich.
<br><br>--
<br>Richard Jones
<br>Red Hat
<br><p>From forum: <a href="http://www.nabble.com/Ocaml-Beginner-f14002.html" embed="fixTarget[14002]" target="_top" >Ocaml Beginner</a></p>tag:www.nabble.com,2006:post-19999418Re: "ocaml_beginners"::[] Simple question, I hope?2008-10-15T11:27:26Z2008-10-15T11:27:26ZFlorent Monnier<div class='shrinkable-quote'>> > I'm writing a simple one dimensional cellular automata using
<br>> > OCaml. I have a space separated file containing 6400 binary bits
<br>> > which I will use as the initial row. How do I include this bit
<br>> > string into OCaml so that it is compiled with the program? I can't
<br>> > have the program parse the file while it is running. Any hints would
<br>> > sure be appreciated.
<br>>
<br>> OCaml doesn't have a native bitstring type, so you can either use a
<br>> (character) string (and either pack the bits or use one per byte) or
<br>> you can use a a bool array (even more wasteful of space, but perhaps
<br>> the most simple and clear).
</div><br>The extLib provides this data structure, see its interface doc:
<br> <a href="http://ocaml-extlib.googlecode.com/svn/doc/apiref/BitSet.html" target="_top" rel="nofollow">http://ocaml-extlib.googlecode.com/svn/doc/apiref/BitSet.html</a><br> <a href="http://code.google.com/p/ocaml-extlib/" target="_top" rel="nofollow">http://code.google.com/p/ocaml-extlib/</a><br><br>--
<br><br>------------------------------------
<br><br>Archives up to December 31, 2007 are also downloadable at <a href="http://www.connettivo.net/cntprojects/ocaml_beginners/" target="_top" rel="nofollow">http://www.connettivo.net/cntprojects/ocaml_beginners/</a><br>The archives of the very official ocaml list (the seniors' one) can be found at <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>Attachments are banned and you're asked to be polite, avoid flames etc.Yahoo! Groups Links
<br><br><*> To visit your group on the web, go to:
<br> <a href="http://groups.yahoo.com/group/ocaml_beginners/" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners/</a><br><br><*> Your email settings:
<br> Individual Email | Traditional
<br><br><*> To change settings online go to:
<br> <a href="http://groups.yahoo.com/group/ocaml_beginners/join" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners/join</a><br> (Yahoo! ID required)
<br><br><*> To change settings via email:
<br> mailto:<a href="http://www.nabble.com/user/SendEmail.jtp?type=post&post=19999418&i=0" target="_top" rel="nofollow">ocaml_beginners-digest@...</a>
<br> mailto:<a href="http://www.nabble.com/user/SendEmail.jtp?type=post&post=19999418&i=1" target="_top" rel="nofollow">ocaml_beginners-fullfeatured@...</a>
<br><br><*> To unsubscribe from this group, send an email to:
<br> <a href="http://www.nabble.com/user/SendEmail.jtp?type=post&post=19999418&i=2" target="_top" rel="nofollow">ocaml_beginners-unsubscribe@...</a>
<br><br><*> Your use of Yahoo! Groups is subject to:
<br> <a href="http://docs.yahoo.com/info/terms/" target="_top" rel="nofollow">http://docs.yahoo.com/info/terms/</a><br><br><p>From forum: <a href="http://www.nabble.com/Ocaml-Beginner-f14002.html" embed="fixTarget[14002]" target="_top" >Ocaml Beginner</a></p>tag:www.nabble.com,2006:post-19999198Re: Quantifier in the type2008-10-15T11:15:03Z2008-10-15T11:15:03ZFlorent Monnier> I put 'a in the interface:
<br>> val fu : 'a -> 'a
<br>> and int in the implementation:
<br>> let fu x = x + 1
<br><br>this interface doesn't reflect the implementation, ocaml will reject it in any
<br>way
<br><br>> So I have universal quantification: for any type 'a function fu can
<br>> consume the argument.
<br><br>No it is not universal, imagine all the types that a user can create, you
<br>cannot handle them all (except maybe with the Obj module, but it's another
<br>question)
<br><br>Anyway if you handle different types, you still have to join them in a single
<br>type :
<br><br>type generic =
<br> | Int of int
<br> | Float of float
<br> | Int64 of int64
<br> ... and so on
<br><br>> So my implementation doesn't comply with that
<br>> universal quatification. And the message I get is following:
<br>>
<br>> Values do not match: val fu : int -> int is not included in val fu : 'a
<br>> -> 'a
<br><br>Yes it doesn't match at all, the type of the implementation is int, nothing
<br>else.
<br><br>> So the declaration of value in mli file is called simply a 'value'. Is
<br>> it intentional?
<br><br>It is just the word in the English language to designate this thing.
<br><br>> I thought that value and it's declaration are separate notions?
<br><br>a value is handle by a identifier and has a type
<br> let ident = some_value in
<br>here ident and some_value are of some type
<br><br>> My reading of " val fu : 'a -> 'a " is:
<br>> some partiular value vvv that belongs to set of values that satisfy
<br>> "forall 'a : (vvv can be used with ('a -> 'a) type)"
<br><br>here are some examples of functions with type ('a -> 'a) :
<br><br>let id v = (v)
<br>let return_sleep v = Unix.sleep 3; (v)
<br>let feedback_kind v =
<br> let r = Obj.repr v in
<br> if Obj.is_int r then print_endline "value similar to int";
<br> (v)
<br>;;
<br><br># let (duplicate : 'a -> 'a) = function v ->
<br> Obj.obj(Obj.dup(Obj.repr v)) ;;
<br>val duplicate : 'a -> 'a = <fun>
<br># let str = "AAA" ;;
<br>val str : string = "AAA"
<br># let str2 = duplicate str ;;
<br>val str2 : string = "AAA"
<br># str.[0] <- 'B' ;;
<br>- : unit = ()
<br># str, str2 ;;
<br>- : string * string = ("BAA", "AAA")
<br><br>________________________________
<br><br><br># let reminder = ref [] ;;
<br>val reminder : '_a list ref = {contents = []}
<br><br># let remind_through v =
<br> if List.exists (fun this -> Obj.magic v = this) !reminder
<br> then failwith "allready processed"
<br> else begin
<br> reminder := Obj.magic v :: !reminder;
<br> (v)
<br> end
<br> ;;
<br>val remind_through : 'a -> 'a = <fun>
<br><br># let a, b, c, d =
<br> "Hello",
<br> 4096,
<br> 4.0 *. atan 1.0,
<br> None;;
<br>val a : string = "Hello"
<br>val b : int = 4096
<br>val c : float = 3.14159265358979312
<br>val d : 'a option = None
<br><br># remind_through a ;;
<br>- : string = "Hello"
<br># remind_through b ;;
<br>- : int = 4096
<br># remind_through c ;;
<br>- : float = 3.14159265358979312
<br># remind_through d ;;
<br>- : 'a option = None
<br><br># remind_through a ;;
<br>Exception: Failure "allready processed".
<br><br><br><br>> But if I write
<br>> let (bar : 'a -> 'a ) = (fun x -> x + 1)
<br>> I create a value that belongs to set "exists 'a : (vvv can be used with
<br>> ('a -> 'a) type)"
<br>> So it's the other quantifier.
<br><br>int can be used for a function which takes 'a, but this opposite is not true:
<br>a function that implements with ints can not take 'a parameters.
<br><br>> I think that the quantifier has to be part of type, since a type is set
<br>> of values (and the quantifier plays important role when describing some
<br>> of such sets).
<br>> So my question is: since we see the same string " 'a -> 'a " that refers
<br>> to different types in different contexts, what are the rules? How is
<br>> type definition in OCaml translated to a type?
<br><br>type definition in OCaml are translated to actual types by inference.
<br><a href="http://en.wiktionary.org/wiki/inference" target="_top" rel="nofollow">http://en.wiktionary.org/wiki/inference</a><br><br>(fun x -> x + 1) will be easily infered to (int -> int)
<br><br>The only way to do a generic function to make addition on generic types as
<br>float, int, or a number in a string, is like the example below.
<br>But those kind of things are HIGHLY discouraged !
<br>Please consider each Obj.magic as a hammer hit,
<br>you will understand that programming with a hammer is discouraged :-)
<br><br># let gen_add v =
<br> let r = Obj.repr v in
<br> let tag = Obj.tag r in
<br> if tag = Obj.int_tag then
<br> (Obj.magic((Obj.magic v) + 1))
<br> else if tag = Obj.double_tag then
<br> (Obj.magic((Obj.magic v) +. 1.0))
<br> else
<br> try if tag = Obj.string_tag then
<br> (Obj.magic(Scanf.sscanf (Obj.magic v) "%f"
<br> (fun v -> string_of_float(v +. 1.0))))
<br> else
<br> (v)
<br> with Scanf.Scan_failure _ -> (v)
<br> ;;
<br>val gen_add : 'a -> 'a = <fun>
<br><br># gen_add 3 ;;
<br>- : int = 4
<br># gen_add 3.0 ;;
<br>- : float = 4.
<br># gen_add "3" ;;
<br>- : string = "4."
<br><br>--
<br>Florent
<br><br>_______________________________________________
<br>Caml-list mailing list. Subscription management:
<br><a href="http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list" target="_top" rel="nofollow">http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list</a><br>Archives: <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>Beginner's list: <a href="http://groups.yahoo.com/group/ocaml_beginners" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners</a><br>Bug reports: <a href="http://caml.inria.fr/bin/caml-bugs" target="_top" rel="nofollow">http://caml.inria.fr/bin/caml-bugs</a><br><p>From forum: <a href="http://www.nabble.com/Caml-Discuss2-f14016.html" embed="fixTarget[14016]" target="_top" >Caml Discuss2</a></p>tag:www.nabble.com,2006:post-19999053Re: OCaml version 3.11.0+beta12008-10-15T11:07:25Z2008-10-15T11:07:25ZRichard Jones-4On Wed, Oct 15, 2008 at 03:40:01PM +0200, Damien Doligez wrote:
<br>> We are pleased to celebrate the birthday of Friedrich Nietzsche
<br>> by releasing OCaml version 3.11.0+beta1. We need YOU to test
<br>> it thoroughly and report any problems you might have. Does
<br>> your favorite software work with it?
<br><br>Excellent news. For Fedora we'll probably leave this until Fedora 11,
<br>which should begin in a week or two.
<br><br>Rich.
<br><br>--
<br>Richard Jones
<br>Red Hat
<br><br>_______________________________________________
<br>Caml-list mailing list. Subscription management:
<br><a href="http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list" target="_top" rel="nofollow">http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list</a><br>Archives: <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>Beginner's list: <a href="http://groups.yahoo.com/group/ocaml_beginners" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners</a><br>Bug reports: <a href="http://caml.inria.fr/bin/caml-bugs" target="_top" rel="nofollow">http://caml.inria.fr/bin/caml-bugs</a><br><p>From forum: <a href="http://www.nabble.com/Caml-Discuss2-f14016.html" embed="fixTarget[14016]" target="_top" >Caml Discuss2</a></p>tag:www.nabble.com,2006:post-19998585Re: "ocaml_beginners"::[] Simple question, I hope?2008-10-15T10:43:07Z2008-10-15T10:43:07ZJon Harrop-3On Wednesday 15 October 2008 18:03:08 Rob Stites wrote:
<br>> I'm writing a simple one dimensional cellular automata using OCaml. I
<br>> have a space separated file containing 6400 binary bits which I will use
<br>> as the initial row. How do I include this bit string into OCaml so
<br>> that it is compiled with the program? I can't have the program parse
<br>> the file while it is running. Any hints would sure be appreciated.
<br>> Thanks.
<br><br>I do this kind of thing all the time: makes it easier for users to deploy the
<br>program.
<br><br>OCaml (particularly ocamlopt) can fall into performance holes during
<br>compilation if you try to do anything non-trivial with real expression so the
<br>best solution I have found is to put the string into the OCaml program as a
<br>verbatim string and then parse it at startup.
<br><br>--
<br>Dr Jon Harrop, Flying Frog Consultancy Ltd.
<br><a href="http://www.ffconsultancy.com/?e" target="_top" rel="nofollow">http://www.ffconsultancy.com/?e</a><br><p>From forum: <a href="http://www.nabble.com/Ocaml-Beginner-f14002.html" embed="fixTarget[14002]" target="_top" >Ocaml Beginner</a></p>tag:www.nabble.com,2006:post-19998562Re: OCaml version 3.11.0+beta12008-10-15T10:41:03Z2008-10-15T10:41:03ZDaniel Bünzli-2<br>Le 15 oct. 08 à 17:27, Andres Varon a écrit :
<br><div class='shrinkable-quote'><br>>>> The native Dynlink is known to work under Linux x86, Linux AMD64,
<br>>>> Win32 (mingw/msvc ports). It has been lightly tested under Win64,
<br>>>> some flavors of BSDs and also the Cygwin port.
<br>>>
<br>>> And on macosx ? It seems here on 10.5.5 that only dynlink.cma and
<br>>> dynlink.cmi for bytecode get installed. So I guess there's no
<br>>> support. What about the future ?
<br>>
<br>> It was installed here: Mac OS X 10.5.5 x86. Maybe this is an issue
<br>> that I mentioned before in the list and got no response?: At least
<br>> one cmxa only get installed if you make opt.opt. camlp4fulllib.cmxa
<br>> was not installed with only make opt (which led to broken
<br>> installations that would not let my apps compile).
</div><br>In fact it was compiled but not installed. I'm using fastword.sh on
<br>10.5.5. x86 and it seems install.sh doesn't install dynlink.cmxa and
<br>dynlink.mli
<br><br>Best,
<br><br>Daniel
<br><br><br>_______________________________________________
<br>Caml-list mailing list. Subscription management:
<br><a href="http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list" target="_top" rel="nofollow">http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list</a><br>Archives: <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>Beginner's list: <a href="http://groups.yahoo.com/group/ocaml_beginners" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners</a><br>Bug reports: <a href="http://caml.inria.fr/bin/caml-bugs" target="_top" rel="nofollow">http://caml.inria.fr/bin/caml-bugs</a><br><p>From forum: <a href="http://www.nabble.com/Caml-Discuss2-f14016.html" embed="fixTarget[14016]" target="_top" >Caml Discuss2</a></p>tag:www.nabble.com,2006:post-19998328Re: "ocaml_beginners"::[] Simple question, I hope?2008-10-15T10:29:13Z2008-10-15T10:29:13ZEric CooperOn Wed, Oct 15, 2008 at 10:03:08AM -0700, Rob Stites wrote:
<br>> I'm writing a simple one dimensional cellular automata using
<br>> OCaml. I have a space separated file containing 6400 binary bits
<br>> which I will use as the initial row. How do I include this bit
<br>> string into OCaml so that it is compiled with the program? I can't
<br>> have the program parse the file while it is running. Any hints would
<br>> sure be appreciated.
<br><br>OCaml doesn't have a native bitstring type, so you can either use a
<br>(character) string (and either pack the bits or use one per byte) or
<br>you can use a a bool array (even more wasteful of space, but perhaps
<br>the most simple and clear).
<br><br>Once you've chosen the representation, you can do something like
<br> $ echo 'let row = "'$(cat datafile)'"' > input.ml
<br>and then compile input.ml together with your program,
<br>which can then reference the value Input.row.
<br><br>You might have to change the simple "echo" to something more complex,
<br>like a shell, awk, or perl script, depending on your representation.
<br><br>--
<br>Eric Cooper e c c @ c m u . e d u
<br><p>From forum: <a href="http://www.nabble.com/Ocaml-Beginner-f14002.html" embed="fixTarget[14002]" target="_top" >Ocaml Beginner</a></p>tag:www.nabble.com,2006:post-19997860"ocaml_beginners"::[] Simple question, I hope?2008-10-15T10:03:08Z2008-10-15T10:03:08ZRob StitesI'm writing a simple one dimensional cellular automata using OCaml. I
<br>have a space separated file containing 6400 binary bits which I will use
<br>as the initial row. How do I include this bit string into OCaml so
<br>that it is compiled with the program? I can't have the program parse
<br>the file while it is running. Any hints would sure be appreciated. Thanks.
<br><br>Rob
<br><p>From forum: <a href="http://www.nabble.com/Ocaml-Beginner-f14002.html" embed="fixTarget[14002]" target="_top" >Ocaml Beginner</a></p>tag:www.nabble.com,2006:post-19997563Re: "ocaml_beginners"::[] What's the names of that construct: let s .... s2008-10-15T09:47:54Z2008-10-15T09:47:54ZFlorent Monnierben.aurel wrote :
<br>> hi
<br>> I'm a OCaml beginner and a beginner to functional programming. So it's
<br>> a bit a lot to learn at once, thats why I like to work with simple
<br>> examples to get a 'feeling' for the language.
<br>>
<br>> Here I've this rather imperative example of reading a file and writing
<br>> its output to another file.
<br><br>Hi!
<br><br>the generic way to load a complete file in a string is to use the Buffer
<br>module, something like the example below. Here we read char by char, but the
<br>IO functions in the Pervasives module are buffered.
<br><br>Also maybe you will like to see simple examples from there:
<br> <a href="http://rosettacode.org/wiki/Category:OCaml" target="_top" rel="nofollow">http://rosettacode.org/wiki/Category:OCaml</a><br><br><br>let read_file_contents filename =
<br> let ic = open_in filename
<br> and buf = Buffer.create 4096 in
<br> try
<br> while true do
<br> let c = input_char ic in
<br> Buffer.add_char buf c; (* fill the buffer *)
<br> done;
<br> assert(false) (* never reached, but for the type coherence *)
<br> with End_of_file ->
<br> close_in ic;
<br> (Buffer.contents buf) (* return *)
<br>;;
<br><br>(* main *)
<br>let () =
<br> let filename =
<br> try Sys.argv.(1)
<br> with _ ->
<br> Printf.eprintf "Usage:\n %s <filename>\n" Sys.argv.(0);
<br> exit 1
<br> in
<br> let contents = read_file_contents filename in
<br> let upper_cont = String.uppercase contents in
<br> (*
<br> (* print it *)
<br> print_string upper_cont;
<br> *)
<br><br> (* output it *)
<br> let output_file = filename ^ ".out" in
<br> let oc = open_out output_file in
<br> output_string oc upper_cont;
<br> flush oc;
<br> close_out oc;
<br>;;
<br><br>--
<br>Florent
<br><br>------------------------------------
<br><br>Archives up to December 31, 2007 are also downloadable at <a href="http://www.connettivo.net/cntprojects/ocaml_beginners/" target="_top" rel="nofollow">http://www.connettivo.net/cntprojects/ocaml_beginners/</a><br>The archives of the very official ocaml list (the seniors' one) can be found at <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>Attachments are banned and you're asked to be polite, avoid flames etc.Yahoo! Groups Links
<br><br><*> To visit your group on the web, go to:
<br> <a href="http://groups.yahoo.com/group/ocaml_beginners/" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners/</a><br><br><*> Your email settings:
<br> Individual Email | Traditional
<br><br><*> To change settings online go to:
<br> <a href="http://groups.yahoo.com/group/ocaml_beginners/join" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners/join</a><br> (Yahoo! ID required)
<br><br><*> To change settings via email:
<br> mailto:<a href="http://www.nabble.com/user/SendEmail.jtp?type=post&post=19997563&i=0" target="_top" rel="nofollow">ocaml_beginners-digest@...</a>
<br> mailto:<a href="http://www.nabble.com/user/SendEmail.jtp?type=post&post=19997563&i=1" target="_top" rel="nofollow">ocaml_beginners-fullfeatured@...</a>
<br><br><*> To unsubscribe from this group, send an email to:
<br> <a href="http://www.nabble.com/user/SendEmail.jtp?type=post&post=19997563&i=2" target="_top" rel="nofollow">ocaml_beginners-unsubscribe@...</a>
<br><br><*> Your use of Yahoo! Groups is subject to:
<br> <a href="http://docs.yahoo.com/info/terms/" target="_top" rel="nofollow">http://docs.yahoo.com/info/terms/</a><br><br><p>From forum: <a href="http://www.nabble.com/Ocaml-Beginner-f14002.html" embed="fixTarget[14002]" target="_top" >Ocaml Beginner</a></p>tag:www.nabble.com,2006:post-19997397Re: Re: Road to native windows OCaml...2008-10-15T09:39:01Z2008-10-15T09:39:01ZXavier Leroy>> Nasm and masm syntaxes differ. You cannot simply interchange them.
<br>>
<br>> What I meant was that caml would generate one more asm syntax, it already
<br>> supports two (masm and gas).
<br><br>... and it's already a major pain, with quite a bit of code
<br>duplication between the masm and gas code emitters. The world doesn't
<br>need yet another symtax for x86 assembly language.
<br><br>- Xavier Leroy
<br><br>_______________________________________________
<br>Caml-list mailing list. Subscription management:
<br><a href="http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list" target="_top" rel="nofollow">http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list</a><br>Archives: <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>Beginner's list: <a href="http://groups.yahoo.com/group/ocaml_beginners" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners</a><br>Bug reports: <a href="http://caml.inria.fr/bin/caml-bugs" target="_top" rel="nofollow">http://caml.inria.fr/bin/caml-bugs</a><br><p>From forum: <a href="http://www.nabble.com/Caml-Discuss2-f14016.html" embed="fixTarget[14016]" target="_top" >Caml Discuss2</a></p>tag:www.nabble.com,2006:post-19997147Re: Native win32 OCaml: recap2008-10-15T09:27:37Z2008-10-15T09:27:37ZKuba OberOn Wednesday 15 October 2008, Kuba Ober wrote:
<br><br>> 6. ocamlopt can use either ml or masm for assembler; ml comes with
<br>> recent Visual Studios. When ml is not present, it would be good
<br>> to have it use nasm instead.
<br><br>Replying to myself: ml == masm, and ml is present in all VS 2008 SP1,
<br>including Express, so no problems there and no need for nasm.
<br><br>Cheers, Kuba
<br><br>_______________________________________________
<br>Caml-list mailing list. Subscription management:
<br><a href="http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list" target="_top" rel="nofollow">http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list</a><br>Archives: <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>Beginner's list: <a href="http://groups.yahoo.com/group/ocaml_beginners" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners</a><br>Bug reports: <a href="http://caml.inria.fr/bin/caml-bugs" target="_top" rel="nofollow">http://caml.inria.fr/bin/caml-bugs</a><br><p>From forum: <a href="http://www.nabble.com/Caml-Discuss2-f14016.html" embed="fixTarget[14016]" target="_top" >Caml Discuss2</a></p>tag:www.nabble.com,2006:post-19997106Re: Re: Road to native windows OCaml...2008-10-15T09:26:04Z2008-10-15T09:26:04ZKuba Ober<div class='shrinkable-quote'>> >> >> All you need is "cl", "ml" and "link" I think (all are MSVC tools).
<br>> >> >
<br>> >> > And you need masm too, right?
<br>> >>
<br>> >> "ml" is just that masm. It's included into MS Visual Studio
<br>> >> Professional edition and up. For Standard edition and below there is
<br>> >> free www.masm32.com.
<br>> >
<br>> > masm32 has an absolutely horrendous license -- have you read it? It's
<br>> > crazy.
<br>>
<br>> OK, I was wrong: masm (ml.exe) is already included into all non-free
<br>> Visual Studio 2008 editions and into free Express edition since 2008
<br>> SP1. So now it's not a problem at all.
<br>>
<br>> > ml of course should be used if available, but otherwise nasm is there and
<br>> > is what should be used whenver masm is called for and unavailable.
<br>>
<br>> Nasm and masm syntaxes differ. You cannot simply interchange them.
</div><br>What I meant was that caml would generate one more asm syntax, it already
<br>supports two (masm and gas). Any differences are surely easy to code.
<br>Since masm is now available and supported, that became a non issue; the
<br>windows platform notes have to be updated to reflect that.
<br><br>Cheers, Kuba
<br><br>_______________________________________________
<br>Caml-list mailing list. Subscription management:
<br><a href="http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list" target="_top" rel="nofollow">http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list</a><br>Archives: <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>Beginner's list: <a href="http://groups.yahoo.com/group/ocaml_beginners" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners</a><br>Bug reports: <a href="http://caml.inria.fr/bin/caml-bugs" target="_top" rel="nofollow">http://caml.inria.fr/bin/caml-bugs</a><br><p>From forum: <a href="http://www.nabble.com/Caml-Discuss2-f14016.html" embed="fixTarget[14016]" target="_top" >Caml Discuss2</a></p>tag:www.nabble.com,2006:post-19996669Re: OCaml version 3.11.0+beta12008-10-15T09:03:32Z2008-10-15T09:03:32ZXavier Leroy<div class='shrinkable-quote'>>> Native dynlink used to work on Mac OS X < 10.5 (x86 only). The new
<br>>> linker in 10.5 does not support linking shared libraries with non-PIC
<br>>> code. It is still possible to use the old linker, called ld_classic,
<br>>> but some libraries (like X11) does not work, so this has been disabled
<br>>> in the configure script.
<br>>>
<br>>> The clean solution to make natdynlink work on recent Mac OS X systems
<br>>> (beside convincing Apple to support the old behavior of their linker
<br>>> in their new implementation) is to change OCaml's x86 backend so that
<br>>> it produces only PIC code (this has been done for the AMD64 port). I
<br>>> don't think there is currently any plan to work on that.
<br>>
<br>> Ouch, this makes it almost a dead end for us. I can offer some time to
<br>> help in this effort, working in the port, or providing feedback. The
<br>> native dynlink and toplevel are, at least to me, the killer features in
<br>> 3.11, but adding another hole for Mac OS X intel (in addition to not
<br>> supporting x86_64) does not seem like the best choice for an
<br>> increasingly popular architecture.
</div><br>Well, we'd very much like to support native dynlink on OS X 10.5,
<br>but Apple is not helping in the least by crippling their linker
<br>compared with the one in 10.4. If anyone from Apple is on this list,
<br>feel free to contact us at <a href="http://www.nabble.com/user/SendEmail.jtp?type=post&post=19996669&i=0" target="_top" rel="nofollow">caml-devel@...</a> for more
<br>information on this regression.
<br><br>- Xavier Leroy
<br><br>_______________________________________________
<br>Caml-list mailing list. Subscription management:
<br><a href="http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list" target="_top" rel="nofollow">http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list</a><br>Archives: <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>Beginner's list: <a href="http://groups.yahoo.com/group/ocaml_beginners" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners</a><br>Bug reports: <a href="http://caml.inria.fr/bin/caml-bugs" target="_top" rel="nofollow">http://caml.inria.fr/bin/caml-bugs</a><br><p>From forum: <a href="http://www.nabble.com/Caml-Discuss2-f14016.html" embed="fixTarget[14016]" target="_top" >Caml Discuss2</a></p>tag:www.nabble.com,2006:post-19996540Re: OCaml version 3.11.0+beta12008-10-15T08:56:44Z2008-10-15T08:56:44ZAndres Varon<br>On Oct 15, 2008, at 11:32 AM, Alain Frisch wrote:
<br><div class='shrinkable-quote'><br>> Daniel Bünzli wrote:
<br>>> And on macosx ? It seems here on 10.5.5 that only dynlink.cma and
<br>>> dynlink.cmi for bytecode get installed. So I guess there's no
<br>>> support. What about the future ?
<br>>
<br>> Native dynlink used to work on Mac OS X < 10.5 (x86 only). The new
<br>> linker in 10.5 does not support linking shared libraries with non-
<br>> PIC code. It is still possible to use the old linker, called
<br>> ld_classic, but some libraries (like X11) does not work, so this has
<br>> been disabled in the configure script.
</div><br>>
<br>>
<br>> The clean solution to make natdynlink work on recent Mac OS X systems
<br>> (beside convincing Apple to support the old behavior of their linker
<br>> in their new implementation) is to change OCaml's x86 backend so
<br>> that it produces only PIC code (this has been done for the AMD64
<br>> port). I don't think there is currently any plan to work on that.
<br><br>Ouch, this makes it almost a dead end for us. I can offer some time to
<br>help in this effort, working in the port, or providing feedback. The
<br>native dynlink and toplevel are, at least to me, the killer features
<br>in 3.11, but adding another hole for Mac OS X intel (in addition to
<br>not supporting x86_64) does not seem like the best choice for an
<br>increasingly popular architecture.
<br><br>Again, thanks for the good work!
<br><br>Andres
<br><div class='shrinkable-quote'><br>>
<br>>
<br>> Alain
<br>>
<br>> _______________________________________________
<br>> Caml-list mailing list. Subscription management:
<br>> <a href="http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list" target="_top" rel="nofollow">http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list</a><br>> Archives: <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>> Beginner's list: <a href="http://groups.yahoo.com/group/ocaml_beginners" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners</a><br>> Bug reports: <a href="http://caml.inria.fr/bin/caml-bugs" target="_top" rel="nofollow">http://caml.inria.fr/bin/caml-bugs</a></div><br>_______________________________________________
<br>Caml-list mailing list. Subscription management:
<br><a href="http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list" target="_top" rel="nofollow">http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list</a><br>Archives: <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>Beginner's list: <a href="http://groups.yahoo.com/group/ocaml_beginners" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners</a><br>Bug reports: <a href="http://caml.inria.fr/bin/caml-bugs" target="_top" rel="nofollow">http://caml.inria.fr/bin/caml-bugs</a><br><p>From forum: <a href="http://www.nabble.com/Caml-Discuss2-f14016.html" embed="fixTarget[14016]" target="_top" >Caml Discuss2</a></p>tag:www.nabble.com,2006:post-19996319Re: OCaml version 3.11.0+beta12008-10-15T08:46:04Z2008-10-15T08:46:04ZAlain Frisch-2Andres Varon wrote:
<br>> OK. Would you recommend that configure scripts use this test to verify
<br>> if the functionality is supported?
<br><br>Yes.
<br><br>-- Alain
<br><br>_______________________________________________
<br>Caml-list mailing list. Subscription management:
<br><a href="http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list" target="_top" rel="nofollow">http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list</a><br>Archives: <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>Beginner's list: <a href="http://groups.yahoo.com/group/ocaml_beginners" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners</a><br>Bug reports: <a href="http://caml.inria.fr/bin/caml-bugs" target="_top" rel="nofollow">http://caml.inria.fr/bin/caml-bugs</a><br><p>From forum: <a href="http://www.nabble.com/Caml-Discuss2-f14016.html" embed="fixTarget[14016]" target="_top" >Caml Discuss2</a></p>tag:www.nabble.com,2006:post-19996300Re: OCaml version 3.11.0+beta12008-10-15T08:44:48Z2008-10-15T08:44:48ZAndres Varon<br>On Oct 15, 2008, at 11:37 AM, Alain Frisch wrote:
<br><br>> Andres Varon wrote:
<br>>> One more question: is it always compiled? or is dynlink.cmxa simply
<br>>> not available in some architectures? if yes, what are those?
<br>>
<br>> As far as I can tell, dynlink.cmxa is always compiled. You will get
<br>> error when "ocamlopt -shared" on those architecture where natdynlink
<br>> is not supported.
<br><br>OK. Would you recommend that configure scripts use this test to verify
<br>if the functionality is supported?
<br><br>Andres
<br><br><br><br>>
<br>>
<br>> -- Alain
<br>>
<br><br>_______________________________________________
<br>Caml-list mailing list. Subscription management:
<br><a href="http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list" target="_top" rel="nofollow">http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list</a><br>Archives: <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>Beginner's list: <a href="http://groups.yahoo.com/group/ocaml_beginners" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners</a><br>Bug reports: <a href="http://caml.inria.fr/bin/caml-bugs" target="_top" rel="nofollow">http://caml.inria.fr/bin/caml-bugs</a><br><p>From forum: <a href="http://www.nabble.com/Caml-Discuss2-f14016.html" embed="fixTarget[14016]" target="_top" >Caml Discuss2</a></p>tag:www.nabble.com,2006:post-19996144Re: OCaml version 3.11.0+beta12008-10-15T08:37:53Z2008-10-15T08:37:53ZAlain Frisch-2Andres Varon wrote:
<br>> One more question: is it always compiled? or is dynlink.cmxa simply not
<br>> available in some architectures? if yes, what are those?
<br><br>As far as I can tell, dynlink.cmxa is always compiled. You will get
<br>error when "ocamlopt -shared" on those architecture where natdynlink is
<br>not supported.
<br><br>-- Alain
<br><br>_______________________________________________
<br>Caml-list mailing list. Subscription management:
<br><a href="http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list" target="_top" rel="nofollow">http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list</a><br>Archives: <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>Beginner's list: <a href="http://groups.yahoo.com/group/ocaml_beginners" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners</a><br>Bug reports: <a href="http://caml.inria.fr/bin/caml-bugs" target="_top" rel="nofollow">http://caml.inria.fr/bin/caml-bugs</a><br><p>From forum: <a href="http://www.nabble.com/Caml-Discuss2-f14016.html" embed="fixTarget[14016]" target="_top" >Caml Discuss2</a></p>tag:www.nabble.com,2006:post-19996034Re: OCaml version 3.11.0+beta12008-10-15T08:32:35Z2008-10-15T08:32:35ZAlain Frisch-2Daniel Bünzli wrote:
<br>> And on macosx ? It seems here on 10.5.5 that only dynlink.cma and
<br>> dynlink.cmi for bytecode get installed. So I guess there's no support.
<br>> What about the future ?
<br><br>Native dynlink used to work on Mac OS X < 10.5 (x86 only). The new
<br>linker in 10.5 does not support linking shared libraries with non-PIC
<br>code. It is still possible to use the old linker, called ld_classic, but
<br>some libraries (like X11) does not work, so this has been disabled in
<br>the configure script.
<br><br>The clean solution to make natdynlink work on recent Mac OS X systems
<br>(beside convincing Apple to support the old behavior of their linker in
<br>their new implementation) is to change OCaml's x86 backend so that it
<br>produces only PIC code (this has been done for the AMD64 port). I don't
<br>think there is currently any plan to work on that.
<br><br>Alain
<br><br>_______________________________________________
<br>Caml-list mailing list. Subscription management:
<br><a href="http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list" target="_top" rel="nofollow">http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list</a><br>Archives: <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>Beginner's list: <a href="http://groups.yahoo.com/group/ocaml_beginners" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners</a><br>Bug reports: <a href="http://caml.inria.fr/bin/caml-bugs" target="_top" rel="nofollow">http://caml.inria.fr/bin/caml-bugs</a><br><p>From forum: <a href="http://www.nabble.com/Caml-Discuss2-f14016.html" embed="fixTarget[14016]" target="_top" >Caml Discuss2</a></p>tag:www.nabble.com,2006:post-19996019Re: OCaml version 3.11.0+beta12008-10-15T08:32:00Z2008-10-15T08:32:00ZJaap BoenderOn Wednesday 15 October 2008 15:40:01 Damien Doligez wrote:
<br>> We are pleased to celebrate the birthday of Friedrich Nietzsche
<br>> by releasing OCaml version 3.11.0+beta1. We need YOU to test
<br>> it thoroughly and report any problems you might have. Does
<br>> your favorite software work with it?
<br><br>For ease of testing, I've very quickly updated the FreeBSD port (lightly
<br>tested, as in "it seems to work for me"). It is available at
<br><a href="http://www.pps.jussieu.fr/~boender/ocamlport.tar.gz" target="_top" rel="nofollow">http://www.pps.jussieu.fr/~boender/ocamlport.tar.gz</a> (to be untarred at the
<br>root of your ports tree).
<br><br>Have fun,
<br><br> Jaap Boender
<br>--
<br>On two occasions I have been asked [by members of Parliament!], "Pray, Mr.
<br>Babbage, if you put into the machine wrong figures, will the right answers
<br>come out?" I am not able rightly to apprehend the kind of confusion of
<br>ideas that could provoke such a question.
<br> -- Charles Babbage
<br><br>_______________________________________________
<br>Caml-list mailing list. Subscription management:
<br><a href="http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list" target="_top" rel="nofollow">http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list</a><br>Archives: <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>Beginner's list: <a href="http://groups.yahoo.com/group/ocaml_beginners" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners</a><br>Bug reports: <a href="http://caml.inria.fr/bin/caml-bugs" target="_top" rel="nofollow">http://caml.inria.fr/bin/caml-bugs</a><br><p>From forum: <a href="http://www.nabble.com/Caml-Discuss2-f14016.html" embed="fixTarget[14016]" target="_top" >Caml Discuss2</a></p>tag:www.nabble.com,2006:post-19996004Re: OCaml version 3.11.0+beta12008-10-15T08:29:17Z2008-10-15T08:29:17ZStefano Zacchiroli-3On Wed, Oct 15, 2008 at 05:04:46PM +0200, Alain Frisch wrote:
<br>> Andres Varon wrote:
<br>>> Thanks for the good work. I would like to know exactly what
<br>>> architectures support the native Dynlink? I did not see this
<br>>> information in the release notes.
<br>> The native Dynlink is known to work under Linux x86, Linux AMD64, Win32
<br>> (mingw/msvc ports). It has been lightly tested under Win64, some flavors
<br>> of BSDs and also the Cygwin port.
<br><br>[ Sorry for being lazy to check by myself, but apparently the question
<br> is of wide-interest. ]
<br><br>Is that something that packagers should worry about or not? I mean,
<br>the specific modules/libraries/... are already detected by configure
<br>and not built/installed/... on architectures missing them or should we
<br>(packages) enforce that?
<br><br>Many thanks in advance (and for native Dynlink of course :-))!
<br>Cheers.
<br><br>--
<br>Stefano Zacchiroli -*- PhD in Computer Science \ PostDoc @ Univ. Paris 7
<br>zack@{upsilon.cc,pps.jussieu.fr,debian.org} -<>- <a href="http://upsilon.cc/zack/" target="_top" rel="nofollow">http://upsilon.cc/zack/</a><br>Dietro un grande uomo c'è sempre /oo\ All one has to do is hit the right
<br>uno zaino -- A.Bergonzoni \__/ keys at the right time -- J.S.Bach
<br><br>_______________________________________________
<br>Caml-list mailing list. Subscription management:
<br><a href="http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list" target="_top" rel="nofollow">http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list</a><br>Archives: <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>Beginner's list: <a href="http://groups.yahoo.com/group/ocaml_beginners" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners</a><br>Bug reports: <a href="http://caml.inria.fr/bin/caml-bugs" target="_top" rel="nofollow">http://caml.inria.fr/bin/caml-bugs</a><br><p>From forum: <a href="http://www.nabble.com/Caml-Discuss2-f14016.html" embed="fixTarget[14016]" target="_top" >Caml Discuss2</a></p>tag:www.nabble.com,2006:post-19995970Re: Re: Road to native windows OCaml...2008-10-15T08:29:08Z2008-10-15T08:29:08ZDmitry Bely-2On Wed, Oct 15, 2008 at 6:35 PM, Kuba Ober <<a href="http://www.nabble.com/user/SendEmail.jtp?type=post&post=19995970&i=0" target="_top" rel="nofollow">kuba@...</a>> wrote:
<div class='shrinkable-quote'><br>> On Wednesday 15 October 2008, Dmitry Bely wrote:
<br>>> On Wed, Oct 15, 2008 at 4:35 PM, Kuba Ober <<a href="http://www.nabble.com/user/SendEmail.jtp?type=post&post=19995970&i=1" target="_top" rel="nofollow">kuba@...</a>> wrote:
<br>>> >> For example, if you download and install OCaml MSVC from
<br>>> >> <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a> and you open a MS Visual Studio 2005 MSDOS shell,
<br>>> >> you can perfectly compile a native application (well I have not done it,
<br>>> >> but I will try tomorrow ;-)
<br>>> >>
<br>>> >> All you need is "cl", "ml" and "link" I think (all are MSVC tools).
<br>>> >
<br>>> > And you need masm too, right?
<br>>>
<br>>> "ml" is just that masm. It's included into MS Visual Studio
<br>>> Professional edition and up. For Standard edition and below there is
<br>>> free www.masm32.com.
<br>>
<br>> masm32 has an absolutely horrendous license -- have you read it? It's
<br>> crazy.
</div><br>OK, I was wrong: masm (ml.exe) is already included into all non-free
<br>Visual Studio 2008 editions and into free Express edition since 2008
<br>SP1. So now it's not a problem at all.
<br><br>> ml of course should be used if available, but otherwise nasm is there and
<br>> is what should be used whenver masm is called for and unavailable.
<br><br>Nasm and masm syntaxes differ. You cannot simply interchange them.
<br><br>- Dmitry Bely
<br><br>_______________________________________________
<br>Caml-list mailing list. Subscription management:
<br><a href="http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list" target="_top" rel="nofollow">http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list</a><br>Archives: <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>Beginner's list: <a href="http://groups.yahoo.com/group/ocaml_beginners" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners</a><br>Bug reports: <a href="http://caml.inria.fr/bin/caml-bugs" target="_top" rel="nofollow">http://caml.inria.fr/bin/caml-bugs</a><br><p>From forum: <a href="http://www.nabble.com/Caml-Discuss2-f14016.html" embed="fixTarget[14016]" target="_top" >Caml Discuss2</a></p>tag:www.nabble.com,2006:post-19995935Re: OCaml version 3.11.0+beta12008-10-15T08:27:30Z2008-10-15T08:27:30ZAndres Varon<br>On Oct 15, 2008, at 11:22 AM, Daniel Bünzli wrote:
<br>>
<br>>> The native Dynlink is known to work under Linux x86, Linux AMD64,
<br>>> Win32 (mingw/msvc ports). It has been lightly tested under Win64,
<br>>> some flavors of BSDs and also the Cygwin port.
<br>>
<br>> And on macosx ? It seems here on 10.5.5 that only dynlink.cma and
<br>> dynlink.cmi for bytecode get installed. So I guess there's no
<br>> support. What about the future ?
<br><br>It was installed here: Mac OS X 10.5.5 x86. Maybe this is an issue
<br>that I mentioned before in the list and got no response?: At least one
<br>cmxa only get installed if you make opt.opt. camlp4fulllib.cmxa was
<br>not installed with only make opt (which led to broken installations
<br>that would not let my apps compile).
<br><br>Andres
<br><div class='shrinkable-quote'><br>>
<br>>
<br>> Thanks,
<br>>
<br>> Daniel
<br>>
<br>> _______________________________________________
<br>> Caml-list mailing list. Subscription management:
<br>> <a href="http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list" target="_top" rel="nofollow">http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list</a><br>> Archives: <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>> Beginner's list: <a href="http://groups.yahoo.com/group/ocaml_beginners" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners</a><br>> Bug reports: <a href="http://caml.inria.fr/bin/caml-bugs" target="_top" rel="nofollow">http://caml.inria.fr/bin/caml-bugs</a></div><br>_______________________________________________
<br>Caml-list mailing list. Subscription management:
<br><a href="http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list" target="_top" rel="nofollow">http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list</a><br>Archives: <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>Beginner's list: <a href="http://groups.yahoo.com/group/ocaml_beginners" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners</a><br>Bug reports: <a href="http://caml.inria.fr/bin/caml-bugs" target="_top" rel="nofollow">http://caml.inria.fr/bin/caml-bugs</a><br><p>From forum: <a href="http://www.nabble.com/Caml-Discuss2-f14016.html" embed="fixTarget[14016]" target="_top" >Caml Discuss2</a></p>tag:www.nabble.com,2006:post-19995829Re: OCaml version 3.11.0+beta12008-10-15T08:22:07Z2008-10-15T08:22:07ZDaniel Bünzli-2<br>Le 15 oct. 08 à 16:55, Andres Varon a écrit :
<br><br>> Thanks for the good work. I would like to know exactly what
<br>> architectures support the native Dynlink? I did not see this
<br>> information in the release notes.
<br><br>I was looking for the same information. This should be added to the
<br>release notes.
<br><br>Le 15 oct. 08 à 17:04, Alain Frisch a écrit :
<br><br>> The native Dynlink is known to work under Linux x86, Linux AMD64,
<br>> Win32 (mingw/msvc ports). It has been lightly tested under Win64,
<br>> some flavors of BSDs and also the Cygwin port.
<br><br>And on macosx ? It seems here on 10.5.5 that only dynlink.cma and
<br>dynlink.cmi for bytecode get installed. So I guess there's no support.
<br>What about the future ?
<br><br>Thanks,
<br><br>Daniel
<br><br>_______________________________________________
<br>Caml-list mailing list. Subscription management:
<br><a href="http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list" target="_top" rel="nofollow">http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list</a><br>Archives: <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>Beginner's list: <a href="http://groups.yahoo.com/group/ocaml_beginners" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners</a><br>Bug reports: <a href="http://caml.inria.fr/bin/caml-bugs" target="_top" rel="nofollow">http://caml.inria.fr/bin/caml-bugs</a><br><p>From forum: <a href="http://www.nabble.com/Caml-Discuss2-f14016.html" embed="fixTarget[14016]" target="_top" >Caml Discuss2</a></p>tag:www.nabble.com,2006:post-19995777Re: OCaml version 3.11.0+beta12008-10-15T08:18:48Z2008-10-15T08:18:48ZAndres Varon<br>On Oct 15, 2008, at 11:04 AM, Alain Frisch wrote:
<br><br>> Andres Varon wrote:
<br>>> Thanks for the good work. I would like to know exactly what
<br>>> architectures support the native Dynlink? I did not see this
<br>>> information in the release notes.
<br>>
<br>> The native Dynlink is known to work under Linux x86, Linux AMD64,
<br>> Win32 (mingw/msvc ports). It has been lightly tested under Win64,
<br>> some flavors of BSDs and also the Cygwin port.
<br><br>Great, thanks for the info. Is there a set of tests that I can run? I
<br>would like to see how it works in Mac OS X which is my primary
<br>development environment and the second most important among our users.
<br><br>One more question: is it always compiled? or is dynlink.cmxa simply
<br>not available in some architectures? if yes, what are those?
<br><br>Andres
<br>>
<br>>
<br>> -- Alain
<br>>
<br><br>_______________________________________________
<br>Caml-list mailing list. Subscription management:
<br><a href="http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list" target="_top" rel="nofollow">http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list</a><br>Archives: <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>Beginner's list: <a href="http://groups.yahoo.com/group/ocaml_beginners" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners</a><br>Bug reports: <a href="http://caml.inria.fr/bin/caml-bugs" target="_top" rel="nofollow">http://caml.inria.fr/bin/caml-bugs</a><br><p>From forum: <a href="http://www.nabble.com/Caml-Discuss2-f14016.html" embed="fixTarget[14016]" target="_top" >Caml Discuss2</a></p>tag:www.nabble.com,2006:post-19995504Re: File inclusion with Ocamldoc2008-10-15T08:05:50Z2008-10-15T08:05:50ZNicolas Pouillard-2Excerpts from Dario Teixeira's message of Wed Oct 15 16:43:32 +0200 2008:
<div class='shrinkable-quote'><br>> Hi,
<br>>
<br>> > It works for me with the 3.10.1. I just tried to write a
<br>> > file test.ml, and a file truc.odocl with "Test" in it, and then
<br>> > ran ocamlbuild :
<br>> > ocamlbuild truc.docdic/index.html
<br>> > And it worked. Or maybe I didn't understand what you want?
<br>>
<br>> You're right. I took the Ocamlbuild manual at face value, since
<br>> it states it cannot handle implementation files. Nevertheless,
<br>> I'm still running into some problems with Ocamlbuild+Ocamldoc.
<br>> First, while Ocamldoc will by default merge the comments from .ml
<br>> and .mli files (with precedence given to the .mli file, unless
<br>> -inv-merge-ml-mli is given), when I use Ocamlbuild to generate
<br>> the documentation I notice the following pattern: if only the .ml
<br>> file is present then it's used for generating the documentation;
<br>> If, however, both .ml and .mli are present, then only the .mli
<br>> is taken into account (no merging is done). Second, how does one
<br>> tell Ocamlbuild to include .txt files in the documentation?
</div><br>The manual states that implementation files are not handled because if you
<br>have a .ml and .mli then ocamlbuild will choose the .mli.
<br><br>--
<br>Nicolas Pouillard aka Ertai
<br><br>_______________________________________________
<br>Caml-list mailing list. Subscription management:
<br><a href="http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list" target="_top" rel="nofollow">http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list</a><br>Archives: <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>Beginner's list: <a href="http://groups.yahoo.com/group/ocaml_beginners" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners</a><br>Bug reports: <a href="http://caml.inria.fr/bin/caml-bugs" target="_top" rel="nofollow">http://caml.inria.fr/bin/caml-bugs</a><br><p>From forum: <a href="http://www.nabble.com/Caml-Discuss2-f14016.html" embed="fixTarget[14016]" target="_top" >Caml Discuss2</a></p>tag:www.nabble.com,2006:post-19995473Re: OCaml version 3.11.0+beta12008-10-15T08:04:46Z2008-10-15T08:04:46ZAlain Frisch-2Andres Varon wrote:
<br>> Thanks for the good work. I would like to know exactly what
<br>> architectures support the native Dynlink? I did not see this information
<br>> in the release notes.
<br><br>The native Dynlink is known to work under Linux x86, Linux AMD64, Win32
<br>(mingw/msvc ports). It has been lightly tested under Win64, some flavors
<br>of BSDs and also the Cygwin port.
<br><br>-- Alain
<br><br>_______________________________________________
<br>Caml-list mailing list. Subscription management:
<br><a href="http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list" target="_top" rel="nofollow">http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list</a><br>Archives: <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>Beginner's list: <a href="http://groups.yahoo.com/group/ocaml_beginners" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners</a><br>Bug reports: <a href="http://caml.inria.fr/bin/caml-bugs" target="_top" rel="nofollow">http://caml.inria.fr/bin/caml-bugs</a><br><p>From forum: <a href="http://www.nabble.com/Caml-Discuss2-f14016.html" embed="fixTarget[14016]" target="_top" >Caml Discuss2</a></p>tag:www.nabble.com,2006:post-19995450Quantifier in the type2008-10-15T08:03:20Z2008-10-15T08:03:20ZDawid Toton-2Just an elementary question.
<br><br>I put 'a in the interface:
<br>val fu : 'a -> 'a
<br>and int in the implementation:
<br>let fu x = x + 1
<br><br>So I have universal quantification: for any type 'a function fu can
<br>consume the argument. So my implementation doesn't comply with that
<br>universal quatification. And the message I get is following:
<br><br>Values do not match: val fu : int -> int is not included in val fu : 'a
<br>-> 'a
<br><br>So the declaration of value in mli file is called simply a 'value'. Is
<br>it intentional?
<br>I thought that value and it's declaration are separate notions?
<br><br>My reading of " val fu : 'a -> 'a " is:
<br> some partiular value vvv that belongs to set of values that satisfy
<br>"forall 'a : (vvv can be used with ('a -> 'a) type)"
<br><br>But if I write
<br>let (bar : 'a -> 'a ) = (fun x -> x + 1)
<br>I create a value that belongs to set "exists 'a : (vvv can be used with
<br>('a -> 'a) type)"
<br>So it's the other quantifier.
<br><br>I think that the quantifier has to be part of type, since a type is set
<br>of values (and the quantifier plays important role when describing some
<br>of such sets).
<br>So my question is: since we see the same string " 'a -> 'a " that refers
<br>to different types in different contexts, what are the rules? How is
<br>type definition in OCaml translated to a type?
<br><br>Dawid
<br><br>_______________________________________________
<br>Caml-list mailing list. Subscription management:
<br><a href="http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list" target="_top" rel="nofollow">http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list</a><br>Archives: <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>Beginner's list: <a href="http://groups.yahoo.com/group/ocaml_beginners" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners</a><br>Bug reports: <a href="http://caml.inria.fr/bin/caml-bugs" target="_top" rel="nofollow">http://caml.inria.fr/bin/caml-bugs</a><br><p>From forum: <a href="http://www.nabble.com/Caml-Discuss2-f14016.html" embed="fixTarget[14016]" target="_top" >Caml Discuss2</a></p>tag:www.nabble.com,2006:post-19995280Re: OCaml version 3.11.0+beta12008-10-15T07:55:18Z2008-10-15T07:55:18ZAndres Varon<br>On Oct 15, 2008, at 9:40 AM, Damien Doligez wrote:
<br><br>> Dear OCaml Users,
<br>>
<br>> We are pleased to celebrate the birthday of Friedrich Nietzsche
<br>> by releasing OCaml version 3.11.0+beta1. We need YOU to test
<br>> it thoroughly and report any problems you might have. Does
<br>> your favorite software work with it?
<br><br><br>Thanks for the good work. I would like to know exactly what
<br>architectures support the native Dynlink? I did not see this
<br>information in the release notes.
<br><br>I need to update configure scripts to include it or not in the link
<br>step because dynlink.cmxa is needed by camlp4fulllib.cmxa, but now I
<br>realize that it's not just a matter of having OCaml >= 3.11.
<br><br>many thanks again,
<br><br>Andres
<br><div class='shrinkable-quote'><br>>
<br>>
<br>> It is available as a source release only (plus documentation),
<br>> from this address:
<br>> < <a href="http://caml.inria.fr/pub/distrib/ocaml-3.11/" target="_top" rel="nofollow">http://caml.inria.fr/pub/distrib/ocaml-3.11/</a> >
<br>>
<br>> It is also available from our CVS server at:
<br>> < <a href="http://camlcvs.inria.fr/" target="_top" rel="nofollow">http://camlcvs.inria.fr/</a> >
<br>> Use tag "ocaml3110beta1" to get the beta release, and tag
<br>> "release311" to track the bug fixes between this and the
<br>> final release of 3.11.0.
<br>>
<br>> Have fun and PLEASE send us some feedback, positive or negative.
<br>>
<br>>
<br>> -- The OCaml team.
<br>>
<br>>
<br>> --------------------- Camlp5 HOW-TO ------------------------
<br>>
<br>> Camlp5 version 5.09 does not work with OCaml 3.11.0+beta1 out of the
<br>> box. A new version compatible with OCaml 3.11.0 should be released
<br>> very soon. In the meantime you can use the following commands (in the
<br>> root directory of the Camlp5 5.09 sources) to compile Camlp5 5.09 with
<br>> OCaml 3.11.0+beta1. Note that you will need to provide the path name
<br>> to a copy of the OCaml 3.11.0+beta1 sources at the line labelled
<br>> "HERE".
<br>>
<br>>
<br>> cp -R ocaml_stuff/3.11 ocaml_stuff/3.11.0
<br>> cp ocaml_src/main/ast2pt.ml_3.11 ocaml_src/main/ast2pt.ml_3.11.0
<br>> ed main/ast2pt.ml <<-EOF
<br>> g/OCAML_3_11/s//& OR OCAML_3_11_0/
<br>> wq
<br>> EOF
<br>> ed top/rprint.ml <<-EOF
<br>> g/OCAML_3_11/s//& OR OCAML_3_11_0/
<br>> wq
<br>> EOF
<br>> ./configure --transitional
<br>> make steal OCAML_SRC=<path-to-ocaml-source-dir> # HERE
<br>> make core
<br>> make bootstrap_sources
<br>> ./configure --transitional
<br>> make world.opt
<br>>
<br>> That's all. Now you can "make install" as usual.
<br>>
<br>> _______________________________________________
<br>> Caml-list mailing list. Subscription management:
<br>> <a href="http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list" target="_top" rel="nofollow">http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list</a><br>> Archives: <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>> Beginner's list: <a href="http://groups.yahoo.com/group/ocaml_beginners" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners</a><br>> Bug reports: <a href="http://caml.inria.fr/bin/caml-bugs" target="_top" rel="nofollow">http://caml.inria.fr/bin/caml-bugs</a></div><br>_______________________________________________
<br>Caml-list mailing list. Subscription management:
<br><a href="http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list" target="_top" rel="nofollow">http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list</a><br>Archives: <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>Beginner's list: <a href="http://groups.yahoo.com/group/ocaml_beginners" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners</a><br>Bug reports: <a href="http://caml.inria.fr/bin/caml-bugs" target="_top" rel="nofollow">http://caml.inria.fr/bin/caml-bugs</a><br><p>From forum: <a href="http://www.nabble.com/Caml-Discuss2-f14016.html" embed="fixTarget[14016]" target="_top" >Caml Discuss2</a></p>tag:www.nabble.com,2006:post-19995036Re: File inclusion with Ocamldoc2008-10-15T07:43:32Z2008-10-15T07:43:32ZdarioteixeiraHi,
<br><br>> It works for me with the 3.10.1. I just tried to write a
<br>> file test.ml, and a file truc.odocl with "Test" in it, and then
<br>> ran ocamlbuild :
<br>> ocamlbuild truc.docdic/index.html
<br>> And it worked. Or maybe I didn't understand what you want?
<br><br>You're right. I took the Ocamlbuild manual at face value, since
<br>it states it cannot handle implementation files. Nevertheless,
<br>I'm still running into some problems with Ocamlbuild+Ocamldoc.
<br>First, while Ocamldoc will by default merge the comments from .ml
<br>and .mli files (with precedence given to the .mli file, unless
<br>-inv-merge-ml-mli is given), when I use Ocamlbuild to generate
<br>the documentation I notice the following pattern: if only the .ml
<br>file is present then it's used for generating the documentation;
<br>If, however, both .ml and .mli are present, then only the .mli
<br>is taken into account (no merging is done). Second, how does one
<br>tell Ocamlbuild to include .txt files in the documentation?
<br><br>Thanks again for your attention,
<br>Dario Teixeira
<br><br><br><br><br><br>_______________________________________________
<br>Caml-list mailing list. Subscription management:
<br><a href="http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list" target="_top" rel="nofollow">http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list</a><br>Archives: <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>Beginner's list: <a href="http://groups.yahoo.com/group/ocaml_beginners" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners</a><br>Bug reports: <a href="http://caml.inria.fr/bin/caml-bugs" target="_top" rel="nofollow">http://caml.inria.fr/bin/caml-bugs</a><br><p>From forum: <a href="http://www.nabble.com/Caml-Discuss2-f14016.html" embed="fixTarget[14016]" target="_top" >Caml Discuss2</a></p>tag:www.nabble.com,2006:post-19995007Native win32 OCaml: recap2008-10-15T07:42:09Z2008-10-15T07:42:09ZKuba OberHere's the recap from the discussion so far. I'm only considering
<br>win32 platform, nothing else. Please pitch in if I didn't get it
<br>right this time:
<br><br>1. OCaml 3.11 will have a non-replaying bytecode debugger that runs
<br>on native ports (built with msvc or mingw).
<br>2. OCaml requires an installed C compiler, linker and assmebler
<br>-- same ones as used during build - to link with native code.
<br>3. Bytecode OCaml requires no external tools to "compile" and run
<br>OCaml code that does not call upon newly defined native functions.
<br>4. ocamlopt does require assembler and linker to produce the executable.
<br>5. OCaml requires at least msys to build itself using either msvc or mingw
<br>compilers. msys is provides bash, make and friends. OCaml does not
<br>build using nmake.
<br>6. ocamlopt can use either ml or masm for assembler; ml comes with
<br>recent Visual Studios. When ml is not present, it would be good
<br>to have it use nasm instead.
<br><br>Cheers, Kuba
<br><br>_______________________________________________
<br>Caml-list mailing list. Subscription management:
<br><a href="http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list" target="_top" rel="nofollow">http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list</a><br>Archives: <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>Beginner's list: <a href="http://groups.yahoo.com/group/ocaml_beginners" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners</a><br>Bug reports: <a href="http://caml.inria.fr/bin/caml-bugs" target="_top" rel="nofollow">http://caml.inria.fr/bin/caml-bugs</a><br><p>From forum: <a href="http://www.nabble.com/Caml-Discuss2-f14016.html" embed="fixTarget[14016]" target="_top" >Caml Discuss2</a></p>tag:www.nabble.com,2006:post-19994885Re: Re: Road to native windows OCaml...2008-10-15T07:35:34Z2008-10-15T07:35:34ZKuba OberOn Wednesday 15 October 2008, Dmitry Bely wrote:
<div class='shrinkable-quote'><br>> On Wed, Oct 15, 2008 at 4:35 PM, Kuba Ober <<a href="http://www.nabble.com/user/SendEmail.jtp?type=post&post=19994885&i=0" target="_top" rel="nofollow">kuba@...</a>> wrote:
<br>> >> For example, if you download and install OCaml MSVC from
<br>> >> <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a> and you open a MS Visual Studio 2005 MSDOS shell,
<br>> >> you can perfectly compile a native application (well I have not done it,
<br>> >> but I will try tomorrow ;-)
<br>> >>
<br>> >> All you need is "cl", "ml" and "link" I think (all are MSVC tools).
<br>> >
<br>> > And you need masm too, right?
<br>>
<br>> "ml" is just that masm. It's included into MS Visual Studio
<br>> Professional edition and up. For Standard edition and below there is
<br>> free www.masm32.com.
</div><br>masm32 has an absolutely horrendous license -- have you read it? It's
<br>crazy.
<br><br>ml of course should be used if available, but otherwise nasm is there and
<br>is what should be used whenver masm is called for and unavailable.
<br><br>Cheers, Kuba
<br><br>_______________________________________________
<br>Caml-list mailing list. Subscription management:
<br><a href="http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list" target="_top" rel="nofollow">http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list</a><br>Archives: <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>Beginner's list: <a href="http://groups.yahoo.com/group/ocaml_beginners" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners</a><br>Bug reports: <a href="http://caml.inria.fr/bin/caml-bugs" target="_top" rel="nofollow">http://caml.inria.fr/bin/caml-bugs</a><br><p>From forum: <a href="http://www.nabble.com/Caml-Discuss2-f14016.html" embed="fixTarget[14016]" target="_top" >Caml Discuss2</a></p>tag:www.nabble.com,2006:post-19994814Re: Road to native windows OCaml...2008-10-15T07:32:55Z2008-10-15T07:32:55ZKuba OberOn Wednesday 15 October 2008, Seo Sanghyeon wrote:
<br>> 2008/10/15 Kuba Ober <<a href="http://www.nabble.com/user/SendEmail.jtp?type=post&post=19994814&i=0" target="_top" rel="nofollow">kuba@...</a>>:
<br>> > For ia64, I don't care much, and if DDK provides masm-x64 then that's
<br>> > probably good enough.
<br>>
<br>> x64 usually means x86_64, not ia64.
<br><br>Even so, 64 bit is out of my scope at this moment.
<br><br>Cheers, Kuba
<br><br>_______________________________________________
<br>Caml-list mailing list. Subscription management:
<br><a href="http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list" target="_top" rel="nofollow">http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list</a><br>Archives: <a href="http://caml.inria.fr" target="_top" rel="nofollow">http://caml.inria.fr</a><br>Beginner's list: <a href="http://groups.yahoo.com/group/ocaml_beginners" target="_top" rel="nofollow">http://groups.yahoo.com/group/ocaml_beginners</a><br>Bug reports: <a href="http://caml.inria.fr/bin/caml-bugs" target="_top" rel="nofollow">http://caml.inria.fr/bin/caml-bugs</a><br><p>From forum: <a href="http://www.nabble.com/Caml-Discuss2-f14016.html" embed="fixTarget[14016]" target="_top" >Caml Discuss2</a></p>