gildor changed the topic of #ocaml to: Discussions about the OCaml programming language | http://caml.inria.fr/ | OCaml 3.12.0 http://bit.ly/aNZBUp
jeddhaberstro has quit [Quit: jeddhaberstro]
pkl has quit [Remote host closed the connection]
sepp2k has quit [Ping timeout: 272 seconds]
sepp2k has joined #ocaml
jeddhaberstro has joined #ocaml
ftrvxmtrx has quit [Ping timeout: 264 seconds]
ftrvxmtrx has joined #ocaml
sepp2k has quit [Quit: Leaving.]
verte has joined #ocaml
jakedouglas has quit [Quit: Leaving.]
mjonsson has joined #ocaml
groovy2shoes has left #ocaml []
verte has quit [Quit: Lost terminal]
verte has joined #ocaml
jeddhaberstro has quit [Quit: jeddhaberstro]
_unK has quit [Remote host closed the connection]
JimmyRcom has quit [Ping timeout: 272 seconds]
jakedouglas has joined #ocaml
jeddhaberstro has joined #ocaml
jakedouglas has quit [Quit: Leaving.]
ftrvxmtrx has quit [Ping timeout: 252 seconds]
ftrvxmtrx has joined #ocaml
ygrek has joined #ocaml
mjonsson has quit [Ping timeout: 276 seconds]
jeddhaberstro has quit [Quit: jeddhaberstro]
Snark has joined #ocaml
Amorphous has quit [Ping timeout: 240 seconds]
boscop_ has left #ocaml []
boscop has joined #ocaml
Amorphous has joined #ocaml
ygrek has quit [Ping timeout: 245 seconds]
grai has joined #ocaml
ikaros has joined #ocaml
vpalle has joined #ocaml
pikachuyann has joined #ocaml
vpalle has quit [Ping timeout: 265 seconds]
Yoric has joined #ocaml
Yoric has quit [Quit: Yoric]
Yoric has joined #ocaml
pesadilla has left #ocaml []
Yoric has quit [Quit: Yoric]
sepp2k has joined #ocaml
JimmyRcom has joined #ocaml
teiresias has joined #ocaml
ulfdoz_ has joined #ocaml
ulfdoz has quit [Ping timeout: 245 seconds]
ygrek has joined #ocaml
<gildor> ygrek: ping
<gildor> git pull .../extutil
<ygrek> re
<gildor> git pull extunix
<gildor> (sorry)
<ygrek> yes, rationale is needed for the library
<ygrek> concerning missing/undefined functions - this is precisely the point, if the program is OS-dependent - better see it at compile time. Anyway, there is also "full" which raises invalid_arg if function is missing - exactly like Unix, so everybody can choose what he likes
<gildor> there are some syscall like setgroups/getgroups that seem reasonnable to add
<gildor> ygrek: well, the changing sig of the module is sorta problematic
<ygrek> and I am thinking of adding yet another orthogonal dimension - multi/single-threaded - e.g. Unix.read/write can be made much more efficient in ST case
<ygrek> gildor, why?
<gildor> because a library built on a system, that don't require "eventfd", will not be buildable on a system without it
<gildor> if the sig only covers what is really needed by the rev dep, it won't be a problem, but here the sig covers everything
<gildor> basically, this kind of changing sig, will prevent using it for building other libraries
<gildor> i.e. extunix.notfull will be used only by programm
<ygrek> yes
<gildor> + it needs pa_have, i.e. extra works
<gildor> I would prefer something more simple,
<gildor> i.e. #ifndef HAVE_XXX caml_raise(XXX); #endif in the C code
<ygrek> I see. Actually having a notfull module was one of the principal targets
<gildor> ygrek: let think about it a little longer, we will see if it is worth or not
<gildor> ygrek: I propose to mark it as "experimental", we will see if it is worth the effort after 1 or 2 release of extunix
<ygrek> sounds reasonable
<ygrek> at least I was motivated when porting some code from nix to windows and getting invalid_arg surprises in runtime
<gildor> there is other way to enforce this ;-)
<gildor> ygrek, some bugs we could look at
<gildor> http://caml.inria.fr/mantis/view.php?id=2533 (<- very interesting)
<ygrek> yes, seen that. Getting dep on type-conv and sexp for fsync is not very good.
<ygrek> core has good io bindings (bigstrings)
<ygrek> for performance
<gildor> http://caml.inria.fr/mantis/view.php?id=480 (the sysconf call is a must have)
<gildor> I'll add this bug to TODO
<gildor> most of them come with code
<ygrek> there is also netsys but when I needed it was not usable on windows out of the box (immediate assert at startup)
<gildor> janest-core seems pretty good, but the documentation is quite complex to navigate
oriba has joined #ocaml
<gildor> and some (not all) features are not portable, making the whole stack uncertain about portability
<gildor> my guest arrives in 30min and I have to do a release of cryptokit
<gildor> ygrek: see you
<sgnb> gildor: setgroups/getgroups/initgroups were added in 3.12 (or before)
<gildor> sgnb: indeed, I just see it in the BTS of INRIA, thanks for the patch
<ygrek> also, one more interesting question - usage/inclusion of not exported functions from caml runtime - say stat_aux in atfile.c
<gildor> if they are not exported, and the license permit it, copy it
<gildor> just like you do
<ygrek> yes, but the type may change (?)
<gildor> -> unit testing ;-)
<ygrek> encode everything in types and say no to testing! :)
<gildor> unfortunatly, we cannot encode everything in types ;-)
<gildor> need to wait another 20 years before creating a fully safe PL
<gildor> (if we are able to kill PHP/Java/COBOL/Pascal in the next years)
<sgnb> COBOL and Pascal are not already dead?
<gildor> sgnb: you don't even imagine how they are alive
<gildor> your bank account is still manage by COBOL
<sgnb> I mean, do people start new projects using these?
<gildor> yes
<sgnb> I understand that there is an existing large codebase
<gildor> everyday
<sgnb> but are people willfully writing COBOL?
<gildor> slavery has been forbidden in France, so I think they are willfully doing this
<sgnb> but it doesn't mean they wouldn't do something else if they could :-)
<gildor> ... there is a lot of people around that will learn at most 1 PL in their live
<gildor> and stick with it for the rest of it
<gildor> this especially true with widely used PL like Java, COBOL or C++
<gildor> if you want to program for the rest of your life, you can
<gildor> this is how the system cannot adopt new PL
ftrvxmtrx has quit [Ping timeout: 240 seconds]
ftrvxmtrx has joined #ocaml
<adrien> flux: btw, gcc-xml is actually still maintained, but you have to use cvs: no tarball
<adrien> gildor: you have a small error in oasis' configure: the descriptions for --enable-gettext and --disable-gettext are the same (
sepp2k has quit [Quit: Leaving.]
<adrien> gildor: with latest ocaml-fileutils and oasis, I get "Unbound value FileUtil.pwd" when building oasis
<adrien> (with ocaml-fileutils 0.4.1 and oasis darcs actually)
<kaustuv_> Do any of the extensions of stdlib's Unix add support for sysconf(3)?
<adrien> gildor: argh, forget that last comment about FileUtil.pwd, it was 100% pebkac, sorry
<kaustuv_> I guess jane st.'s stdlib replacement has it...
kaustuv_ is now known as kaustuv
ftrvxmtrx has quit [Ping timeout: 240 seconds]
ftrvxmtrx has joined #ocaml
psnively has joined #ocaml
<thelema> I see a batteries package in oasis/dev/browse, but don't know what I should do to help this along
bohanlon has joined #ocaml
ygrek has quit [Ping timeout: 245 seconds]
mjonsson has joined #ocaml
Gentooer has joined #ocaml
vpalle has joined #ocaml
ygrek has joined #ocaml
psnively has quit [Quit: psnively]
Anarchos has joined #ocaml
_unK has joined #ocaml
ulfdoz_ has quit [Quit: Reconnecting]
ulfdoz has joined #ocaml
verte has quit [Quit: ~~~ Crash in JIT!]
<gildor> thelema: hi, batteries in oasis/dev/browse is an example of tarball without _oasis
<gildor> thelema: in a (near) futur, there will be a "Create _oasis file" button on this web page
<thelema> gildor: okay.
<thelema> very nice
<gildor> thelema: just need to create it ;-)
<gildor> (it won't deal with build system, but will help to define metadata about the package, like homepage, authors, copyright, categories and build deps)
<thelema> That'll get me started.
Associat0r has joined #ocaml
ftrvxmtrx has quit [Read error: Connection reset by peer]
ftrvxmtrx has joined #ocaml
sepp2k has joined #ocaml
Yoric has joined #ocaml
ulfdoz has quit [Ping timeout: 276 seconds]
Yoric has quit [Quit: Yoric]
Yoric has joined #ocaml
vpalle has quit [Ping timeout: 255 seconds]
joewilliams is now known as joewilliams_away
jakedouglas has joined #ocaml
Snark has quit [Quit: Ex-Chat]
vpalle has joined #ocaml
Yoric has quit [Quit: Yoric]
Yoric has joined #ocaml
vpalle has quit [Ping timeout: 265 seconds]
ulfdoz has joined #ocaml
Leonidas has quit [Ping timeout: 260 seconds]
fraggle_ has quit [Remote host closed the connection]
vpalle has joined #ocaml
fraggle_ has joined #ocaml
pikachuyann has quit [Quit: 'nuit/'night]
vpalle_ has joined #ocaml
Gentooer has quit [Read error: Connection reset by peer]
vpalle has quit [Ping timeout: 240 seconds]
Leonidas has joined #ocaml
ygrek has quit [Ping timeout: 245 seconds]
Anarchos has quit [Quit: Vision[0.9.7-H-090423]: i've been blurred!]
Yoric has quit [Quit: Yoric]
boscop_ has joined #ocaml
boscop has quit [Ping timeout: 265 seconds]
ikaros has quit [Quit: Leave the magic to Houdini]
bohanlon has quit [Quit: Farewell!]
Anarchos has joined #ocaml
Anarchos has quit [Quit: Vision[0.9.7-H-090423]: i've been blurred!]
oriba has left #ocaml []
vpalle_ has quit [Ping timeout: 252 seconds]
Associat0r has quit [Quit: Associat0r]
grai has quit [Quit: Changing server]