gmane.comp.lang.caml.inria

Finall Call For Papers (DSL WC)

Emir Pasalic — 2008-12-05T03:18:29

_______________________________________________ Caml-list mailing list. Subscription management: http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list Archives: http://caml.inria.fr Beginner's list: http://groups.yahoo.com/group/ocaml_beginners Bug reports: http://caml.inria.fr/bin/caml-bugs

Re: OCaml version 3.11.0 released.

Rich Neswold — 2008-12-05T03:15:54

Re: Computing with big numbers?

Martin Jambon — 2008-12-04T21:51:54

It's like being able to manufacture a lock and the key that goes with it: big deal! :-) Martin

Re: Computing with big numbers?

David Thomas — 2008-12-04T16:40:22

That depends on the threat model. If the question is, "presuming no active attack, how likely is it to break?", then the cryptanalytic results against the hash are irrelevant. If the question is "how secure is it if someone is maliciously manipulating files", then they are certainly relevant. If you're operating between reasonably secure machines, where an attacker having write access is already more catastrophic than a failure of Unison, then the first is what matters. If someone else has control over some of the files, then you've gotta watch the second. --- On Thu, 12/4/08, Florian Hars bik-gmbh.de> wrote: _______________________________________________ Caml-list mailing list. Subscription management: http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list Archives: http://caml.inria.fr Beginner's list: http://groups.yahoo.com/group/ocaml_beginners Bug reports: http://caml.inria.fr/bin/caml-bugs

Re: Computing with big numbers?

Florian Hars — 2008-12-04T16:06:06

Alan Schmitt schrieb: Then it is *especially* relevant, as it is quite trivial to generate several files with different content and the same MD5 hash, all you need is a Playstation 3: http://www.win.tue.nl/hashclash/Nostradamus/ - Florian

OCaml version 3.11.0 released.

Damien Doligez — 2008-12-04T14:52:37

Dear OCaml users, It is our pleasure to celebrate the 51st birthday of Eric S. Raymond by releasing OCaml version 3.11.0. This release brings you many bug fixes and a few much-wanted features such as dynamic linking in native code (not available on all architectures, though). See below for the complete list of changes from the previous version. It is available here: < http://caml.inria.fr/download.en.html > Please note: at this time it is only available as source and as binary for Mac OS X on Intel processors. Other binary versions will be added to the Web site next week. Happy hacking,

2nd Cfp : PAPP 2009 (Part of ICCS 2009)

Frédéric Gava — 2008-12-03T14:18:49

--------------------------------------------------------------------- Please accept our apologies if you have received multiple copies. Please feel free to distribute it to those who might be interested. --------------------------------------------------------------------- ---------------------------------------------------------- 2nd CALL FOR PAPERS PAPP 2009 Sixth International Workshop on aPpplications of declArative and object-oriented Parallel Programming http://lacl.univ-paris12.fr/gava/PAPP2009/ part of ICCS 2009 The International Conference on Computational Science May 25-27, 2009, Baton Rouge, Louisiana, USA ---------------------------------------------------------- NEW ! ----- New deadline (6/12 ==> 15/12) and maximal number of pages (8 pages LNCS ==> 10 pages) AIMS AND SCOPE Computational Science applications are more and more complex to develop and require more

Re: ocamlbuild rule with recursive action

Daniel Bünzli — 2008-12-03T09:58:23

Le 3 déc. 08 à 10:44, Romain Bardou a écrit : But doesn't this break things ? Actions are supposed to build the command not execute it. Daniel P.S. something like Rec of 'a * ('a -> action -> [ `Stop | `Step of command * 'a ]) would be better or Rec of 'a * ('a -> action-> command * 'a) with the convention that returning Nop stops the recursion. _______________________________________________ Caml-list mailing list. Subscription management: http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list Archives: http://caml.inria.fr Beginner's list: http://groups.yahoo.com/group/ocaml_beginners Bug reports: http://caml.inria.fr/bin/caml-bugs

Re: ocamlbuild rule with recursive action

Romain Bardou — 2008-12-03T09:44:56

Daniel Bünzli a écrit : Well, you can also simply execute the commands yourself, in a loop, using Command.execute or Command.execute_many.

Re: ocamlbuild rule with recursive action

Daniel Bünzli — 2008-12-02T23:51:57

Le 2 déc. 08 à 12:27, DooMeeR a écrit : Thanks. I'll try something along these lines. But that's kind of ugly. In the future could maybe ocamlbuild add another case in the command type, something like Rec of action -> (command * bool) Or maybe there is another way ? Daniel _______________________________________________ Caml-list mailing list. Subscription management: http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list Archives: http://caml.inria.fr Beginner's list: http://groups.yahoo.com/group/ocaml_beginners Bug reports: http://caml.inria.fr/bin/caml-bugs

Re: OCaml 3.11.0 release candidate

Jacques Carette — 2008-12-02T22:27:38

In the CYGWIN32, SUPPORT_DYNAMIC_LINKING version of 3.11.0+rc1, file unix.c has the following lines: 178: return flexdll_dlopen(libname, flags); 193: return flexdll_dlsym(flexdll_dlopen(NULL,0,1), name); My version of gcc is extremely unhappy about line 193 [and says so loudly by quitting], since it corresponds to a call to flexdll_dlopen with an incorrect number of parameters. Jacques _______________________________________________ Caml-list mailing list. Subscription management: http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list Archives: http://caml.inria.fr Beginner's list: http://groups.yahoo.com/group/ocaml_beginners Bug reports: http://caml.inria.fr/bin/caml-bugs

Re: Issues with Sexplib (#2)

Markus Mottl — 2008-12-02T19:21:12

On Sat, Nov 29, 2008 at 12:14 PM, Dario Teixeira yahoo.com> wrote: This should be fixed in newer releases of type-conv. Thanks for the feedback! Regards, Markus

Re: Issues with Sexplib (#1)

Markus Mottl — 2008-12-02T19:14:45

On Sat, Nov 29, 2008 at 11:40 AM, Dario Teixeira yahoo.com> wrote: [snip] This is not just a matter of phantom types. Even ordinary abstract types suffer from this if there are constraints on what constitutes a well-formed value, i.e. not every concrete representation conforms to a valid value. E.g. if some abstract type t represents a sorted list and the internal representation is just a list, somebody could pass you an S-expression corresponding to an unsorted list. The only way around this problem is to have functions that check at runtime whether some value of the internal, concrete type corresponds to a valid abstract type. Once you have implemented such a function (e.g. "check_sorted", etc.), you can always easily wrap the converter for reading from S-expressions to call this test right after deserializing. If possible, the best way to handle such situations is to come up with extensional representations that syntactically only allow construction of legal values. Then you do not n

TSI numérospécial ANALYSE STATIQUE et COMPILATION: extension des envois au 15 février

Sandrine Blazy — 2008-12-02T17:18:31

Re: ocamlbuild rule with recursive action

DooMeeR — 2008-12-02T11:27:05

Daniel Bünzli a écrit : Well, you can define a rule like this : rule ~prod: "%.rec" (fun env build -> let x = env "%2.rec" in (* a new fresh name *) if some condition then build x else Nop) When your rule is executed, for exemple to produce "toto.rec", it will (if some condition) first produce "toto2.rec", which will produce "toto22.rec", and so on until some condition is false. I didn't try it though.

ocamlbuild rule with recursive action

Daniel Bünzli — 2008-12-02T10:40:57

Hello, Is it possible to define a rule with an action that is performed repeatedly until a condition is reached ? Best, Daniel _______________________________________________ Caml-list mailing list. Subscription management: http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list Archives: http://caml.inria.fr Beginner's list: http://groups.yahoo.com/group/ocaml_beginners Bug reports: http://caml.inria.fr/bin/caml-bugs

Re: Computing with big numbers?

Alan Schmitt — 2008-12-01T14:37:07

Re: Computing with big numbers?

Alan Schmitt — 2008-12-01T14:29:21

Re: Computing with big numbers?

Dario Teixeira — 2008-12-01T13:47:22

Hi, I reckon that by saying "how good 128 bits MD5 hashes were" you are aware of the recent attacks that make MD5's effective security less than 128-bit. The Wikipedia has a good summary: http://en.wikipedia.org/wiki/MD5 Cheers, Dario Teixeira _______________________________________________ Caml-list mailing list. Subscription management: http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list Archives: http://caml.inria.fr Beginner's list: http://groups.yahoo.com/group/ocaml_beginners Bug reports: http://caml.inria.fr/bin/caml-bugs

Re: Computing with big numbers?

Martin Jambon — 2008-12-01T12:52:54

If I understand your problem correctly, this is the so-called birthday problem with 2^128 days in a year. The Wikipedia article gives useful approximations: http://en.wikipedia.org/wiki/Birthday_problem Martin

Computing with big numbers?

Alan Schmitt — 2008-12-01T12:36:50