mfp changed the topic of #ocaml to: Discussions about the OCaml programming language | http://caml.inria.fr/ | OCaml 3.11.2 released | Inscription for OCaml Meeting 2010 is opened http://wiki.cocan.org/events/europe/ocamlmeetingparis2010
ido has joined #ocaml
smimou has quit [Ping timeout: 246 seconds]
peper has quit [Ping timeout: 256 seconds]
seanmcl has quit [Quit: seanmcl]
tmaeda is now known as tmaedaZ
BigJ has joined #ocaml
execve has joined #ocaml
pimmhogeling has quit [Ping timeout: 256 seconds]
valross has joined #ocaml
sysop_fb has quit [Read error: Connection reset by peer]
sysop_fb has joined #ocaml
threeve has joined #ocaml
CcSsNET has joined #ocaml
dark has quit [Ping timeout: 264 seconds]
yziquel has joined #ocaml
dark has joined #ocaml
boscop has quit [Quit: Gxis revido!]
valross has quit [Remote host closed the connection]
execve has quit [Ping timeout: 252 seconds]
Associat0r has joined #ocaml
Associat0r has quit [Quit: Associat0r]
maskd has quit [Quit: leaving]
valross has joined #ocaml
threeve has quit [Quit: threeve]
mutew has joined #ocaml
mutew has quit [Client Quit]
CcSsNET has quit [Read error: Connection reset by peer]
CcSsNET has joined #ocaml
thrasibule has joined #ocaml
Yoric has joined #ocaml
ulfdoz has joined #ocaml
Yoric has quit [Quit: Yoric]
Associat0r has joined #ocaml
schme has quit [Ping timeout: 245 seconds]
schme has joined #ocaml
schme has quit [Changing host]
schme has joined #ocaml
Associat0r has quit [Client Quit]
Gooffy has joined #ocaml
yakischloba has quit [Quit: Leaving.]
ulfdoz has quit [Ping timeout: 272 seconds]
jonafan_ has joined #ocaml
cognizant-cog has joined #ocaml
jonafan has quit [Ping timeout: 248 seconds]
_zack has joined #ocaml
Associat0r has joined #ocaml
peper has joined #ocaml
ttamttam has joined #ocaml
<mrvn> moin
<julm> plu
cognizant-cog has quit [Quit: ChatZilla 0.9.86 [Firefox 3.5.7/20100106054534]]
pimmhogeling has joined #ocaml
nop has joined #ocaml
Associat0r has quit [Quit: Associat0r]
nop has quit [Quit: leaving]
_unK has joined #ocaml
yziquel has quit [Ping timeout: 252 seconds]
lena_ has joined #ocaml
CcSsNET has quit [Quit: User disconnected]
rwmjones has joined #ocaml
gim has joined #ocaml
smimou has joined #ocaml
execve has joined #ocaml
_unK has quit [Remote host closed the connection]
valross has quit [Quit: Ex-Chat]
_unK has joined #ocaml
jondoh has joined #ocaml
jondoh has quit [Remote host closed the connection]
<gildor> mrvn: I create the git repo and migrate the data from svn for libaio-ocaml
<mrvn> gildor: thanks. Debian git too?
<gildor> mrvn: before debian git, I need to produce a tarball
<gildor> what version should I use ?
<gildor> 0.1.0 ?
<mrvn> I used 0.0.1. If you change something then it must be higher. anything will do.
<gildor> (I will tag the upstream git accordingly and upload a tarball to forge.ocamlcore.org)
<gildor> I didn't change anything
<gildor> so version = 0.0.1, ok ?
<mrvn> fine by me.
boscop has joined #ocaml
<mrvn> You probably should create a changelog saying initial release or something.
<gildor> ok
f[x] has quit [Remote host closed the connection]
maskd has joined #ocaml
<gildor> mrvn: done: git tag, tarball uploaded to forge.o.o, debian repository setup
<gildor> mrvn: I will have a closer look at the debian packaging tonight
<gildor> mrvn: consider checking out the git of git.ocamlcore.org
<mrvn> Feel free to trow it away. I just cut&pasted some other packages debian dir last year.
<gildor> git clone ssh://git.ocamlcore.org/gitroot/libaio-ocaml/libaio-ocaml.git
<gildor> mrvn: indeed there has been a lot of changes since then
<gildor> mrvn: would you like to upload to experimental or unstable ?
<mrvn> unstable
<mrvn> Whats the url for the debian dir?
<gildor> mrvn: right now, I need to work on some other project, I will come back to libaio tonight
<gildor> mrvn: install dh-ocaml package and dom-git-checkout libaio-ocaml
<gildor> "dom-git-checkout libaio-ocaml" is the command to run to get the debian package
<mrvn> How do I set the user for that?
<gildor> through .ssh/config with Host alioth.debian.org
<mrvn> you mean git.d.o and that doesn't work.
<mrvn> Host alioth.debian.org User goswin-guest
<mrvn> ssh -vvv alioth.debian.org
<gildor> so that must be Host git.debian.org \n User goswin-guest
<mrvn> Password:
<mrvn> Options for port and similar work but the "User" setting is plain ignored.
<gildor> mrvn: BTW, we should switch to channel #debian-ocaml
<mrvn> or #ssh
<dark> there is #debian-ocaml ? o.o
<dark> there isn't lol
<dark> I think that this channel is so low-traffic that some tangential discussion is just fine=D
<mrvn> dark: #debian-ocaml is basically just about packaging ocaml stuff. Better target audience there.
<dark> the channel is empty
<dark> :t
<mrvn> irc.debian.org
<dark> ah lol
<dark> btw why debian moved to oftc? i read some msgs in debian-devel, never understood
<dark> #debian at freenode is actually larger
<dark> [in number of users]
<gildor> mrvn: the User option works for me
<gildor> mrvn: please copy and paste the relevant part of your .ssh/config file to http://debian.pastebin.com/
<gildor> try with a single space at the beginning of line (seems silly, but why not)
peddie has quit [Ping timeout: 276 seconds]
<mrvn> no effect
<gildor> mrvn: can you copy and paste the output of "ssh -v alioth.debian.org"
<mrvn> and now alioth has died.
<gildor> unfortunately not, you have been banned due to too many connection attempt
<mrvn> hehe. or that.
<gildor> is /home/mrvn/.ssh/id_rsa the usual key you are using to connect to alioth ?
<mrvn> Should be.
<gildor> well we need to wait at least 2hours before trying again
<mrvn> I tried this out with local systems running sshd in debug mode and ssh simply connects with the default user instead of the one in .ssh/config.
<gildor> check that /home/mrvn/.ssh/id_rsa.pub match the SSH key preferences in your account on alioth.debian.org
<mrvn> Host chocos User foobar
<mrvn> % ssh chocos
<mrvn> mrvn@chocos:~%
<mrvn> That should fail as I have no user foobar
<gildor> mrvn: I don't know, this is a pure ssh problem, I have never encountered this kind of behavior
<mrvn> If I add "Port 1234" I get ssh: connect to host chocos port 1234: Connection refused
<mrvn> Any option I tried works except User.
<gildor> maybe you have a higher precedence option that override this one ?
<mrvn> Not unless the default /etc/ssh/ssh_config has it set
f[x] has joined #ocaml
<gildor> do you have an alias for ssh command ? an environment variable ?
<mrvn> Hmm, alias could be.
<mrvn> thx. That was it. I had an ssh alias that adds -l user depending on the hostname back from my uni days.
<flux> mrvn, heard of .ssh/config?-)
<mrvn> flux: not 10 years ago
<gildor> mrvn: I let you close bug 570363 against dh-ocaml
<mrvn> gildor: There should still be an option for it.
<mrvn> gildor: or at least a stanza in the manpage telling about .ssh/config
<gildor> gildor: for at most 10 users, with another way of doing it, I doubt anybody will implement it
<mrvn> gildor: Now I only need one more thing: For dpkg-source to cope with debian being a symlink and follow it when generating source.
<gildor> mrvn: why do you want to do a symlink ?
<gildor> ... ok you cannot checkout a subdirectory with git
<mrvn> I have debian/git/ and upstream/git/ and upstream/git/debian -> ../../debian/git/debian
<mrvn> That way I can edit the upstream source and run debuild without having to merge every change into the debian git first.
alinbsp has joined #ocaml
Gooffy has quit [Ping timeout: 264 seconds]
<_zack> has anyone seen thelema around today?
Gooffy has joined #ocaml
th5 has joined #ocaml
Modius has quit [Read error: Connection reset by peer]
Modius has joined #ocaml
Associat0r has joined #ocaml
mrvn has quit [Ping timeout: 256 seconds]
<gildor> anyone know if there a way to order build dependencies in ocamlbuild ?
<gildor> i.e. i have "toto.byte": use_libfoo, use_libbar
<gildor> but the command line should be ocamlc ... libbar.cma libfoo.cma
<gildor> ...
<gildor> because foo depends on bar
tmaedaZ is now known as tmaeda
mrvn has joined #ocaml
<f[x]> gildor, order calls to ocaml_lib in myocamlbuild.ml
<gildor> already done
<f[x]> strange then, show the code
<gildor> f[x]: order of use_* in _tags and order of ocaml_lib doesn't change anything
<f[x]> order of ocaml_lib definitely changes the command line
<gildor> give me 2 sec, checking
<gildor> f[x]: which OCaml version are you using
<gildor> f[x]: here I use 3.11.2 and it doesn't change anything
execve has quit [Quit: leaving]
<f[x]> hm, 3.11.1
<f[x]> can you show the code? I am puzzled
pimmhogeling has quit [Ping timeout: 272 seconds]
<gildor> f[x]: just have a look at ocaml_lib code, it uses an Hashtbl
<gildor> f[x]: you want a tarball of the project ?
<f[x]> tarball/vcs
<gildor> VCS
<gildor> I need to commit my change
<gildor> and the code is generated as a test case
synfin has joined #ocaml
<f[x]> gildor, I believe the order comes from the flags which inject cma into command line, and the flags are kept in global single-linked list
<f[x]> see flags.ml
tmaeda is now known as tmaedaZ
<gildor> will see, I encounter a bug in Scanf with 3.11.2 which doesn't exist before
<gildor> (bugs day it seems)
_zack has quit [Read error: Connection reset by peer]
orbitz has quit [Ping timeout: 256 seconds]
jimmyb2187 has left #ocaml []
_zack has joined #ocaml
<gildor> f[x]: humm, I realize you will have to compile a lot of thing including not-yet-published library
<gildor> I prepare a tarball
<gildor> of the example alone
jimmyb2187 has joined #ocaml
derdon has joined #ocaml
<f[x]> ouch
<gildor> f[x]: a problem ?
<gildor> I am sorry the setup.ml is quite big
<gildor> for such a small example
thelema has joined #ocaml
<gildor> thelema: _zack is looking for you
<thelema> gildor: thank you
<f[x]> gildor, yes, the problem is here, maybe it is because these are "internal" libs..
<gildor> f[x]: honnestly, I think it is a bug, there should be a way to define order, what is your opinion ?
<gildor> f[x]: if you agree, I will submit a bug against ocaml
Modius has quit [Quit: I'm big in Japan]
<f[x]> a moment
eldragon has quit [Read error: Connection reset by peer]
eldragon has joined #ocaml
<f[x]> it discovers lib*.cma as dynamic dependencies and the order is wrong
<f[x]> hm
<f[x]> anyway it is a bug
Associat0r has quit [Quit: Associat0r]
_zack has quit [Quit: Leaving.]
<f[x]> on the clean tree - ocamlbuild src/execa/Main.byte will not find module C, this feels wrong
<f[x]> you are not using include in _tags but specify the dir manually in plugin - yes? probably that confuses ocamlbuild
<gildor> f[x]: indeed I use the ~dir arg of ocaml_lib to define include dir
<gildor> this should work
<gildor> but this has nothing to do with *.cma order ?
tmaedaZ is now known as tmaeda
<f[x]> who knows
chupish has joined #ocaml
<gildor> f[x]: but do you have build module C before calling ocamlbuild src/execa/Main.byte ?
<f[x]> <src/lib{a,b,c}>: include # and everything works fine
<gildor> f[x]: unfortunately this is too general, interdepends is just an example, it should work without a so general include
<f[x]> and then it doesn't depend on ocaml_lib order - and now this feels natural - for internal libraries it is able to derive dependencies and order them easily
<f[x]> gildor, I understand
<gildor> f[x]: I even think that is just fetch directly the .cmi/.cmo/.cmx directly in the directory
Associat0r has joined #ocaml
<gildor> what does the command line look like when compiling execa ?
<f[x]> ocamlfind ocamlc -linkpkg -package unix -package str -package num src/liba/liba.cma src/libb/libb.cma src/libc/libc.cma src/execa/Main.cmo -o src/execa/Main.byte
<gildor> f[x]: so maybe this is just a matter of adding the include dir from tags when computing library order ?
qwr has joined #ocaml
<f[x]> gildor, so then if you don't want to treat them as "internal" libraries from ocamlbuild point of view - threat them as "external"!
<diml> gildor: you can make ocamlbuild to order things correctly by using "dep" (see http://pastebin.com/m6eaad20)
<f[x]> as you are supplying dependency info and willing to guarantee that libs are built before main.byte (with setup.ml)
<f[x]> ocaml_lib ~extern:true ~dir:dir lib;
<f[x]> and everything works fine :)
qwr has quit [Quit: Lost terminal]
|Jedai| has quit [Quit: KVIrc Insomnia 4.0.0, revision: , sources date: 20090520, built on: 2009/06/08 19:18:46 UTC http://www.kvirc.net/]
<f[x]> diml, the question is why doesn't ocaml_lib put such deps? It uses more specific ones only (with "link" tag)..
<gildor> f[x], diml: I think the extern flags is wrongly named
<gildor> f[x], diml: it just explain that it should depends on *.cma|cmxa
<gildor> f[x], diml: an internal library will probably compile directly with cmo/cmx
qwr has joined #ocaml
<mrvn> Every cmo/cmx file contains a list of modules it depends on, right?
<gildor> yes
<mrvn> Then why can't ocamlc/opt sort them on the fly? Can one have different cmo/cmx files with the same module linked together?
_zack has joined #ocaml
avsm has joined #ocaml
tmaeda is now known as tmaedaZ
Gooffy has quit [Quit: Leaving.]
lena_ has quit [Remote host closed the connection]
Alpounet has quit [Quit: ``Do what you think you can't do.'']
yakischloba has joined #ocaml
<gildor> f[x], diml: thanks a lot, it works with extern:true, I also implement a flag to use "libx: include" which set extern to false and add the include stanza in _tags
yakischloba has left #ocaml []
yakischloba has joined #ocaml
th5 has quit [Quit: th5]
ikaros has joined #ocaml
lokydor has joined #ocaml
_zack has quit [Quit: Leaving.]
jonafan_ is now known as jonafan
jcaose has joined #ocaml
chupish has quit []
Modius has joined #ocaml
Modius has quit [Max SendQ exceeded]
Modius has joined #ocaml
yakischloba has quit [Quit: Leaving.]
yakischloba has joined #ocaml
yakischloba has quit [Ping timeout: 260 seconds]
<Camarade_Tux> about cross-compilers in fedora and debian: do they come with additional libraries?
<Camarade_Tux> (especially interested in sexplib)
<Camarade_Tux> actually: mikmatch-pcre, sexplib, ocaml-fileutils
Anarchos has joined #ocaml
_unK has quit [Remote host closed the connection]
ttamttam has quit [Quit: Leaving.]
synfin has quit [Ping timeout: 260 seconds]
yakischloba has joined #ocaml
lokydor has quit [Ping timeout: 252 seconds]
lokydor has joined #ocaml
_unK has joined #ocaml
rwmjones has quit [Ping timeout: 272 seconds]
ulfdoz has joined #ocaml
rwmjones has joined #ocaml
DaveS has quit [Quit: leaving]
fraggle_ has quit [Ping timeout: 248 seconds]
fraggle_ has joined #ocaml
yakischloba has quit [Quit: Leaving.]
rwmjones has quit [Ping timeout: 256 seconds]
bzzbzz has joined #ocaml
alinbsp has quit [Quit: alinbsp]
rwmjones has joined #ocaml
yakischloba has joined #ocaml
Smerdyakov has joined #ocaml
jcaose has quit [Read error: Connection reset by peer]
<mehdid> Camarade_Tux: I don't think so
ulfdoz has quit [Ping timeout: 252 seconds]
ulfdoz has joined #ocaml
<Smerdyakov> Can I recruit any new people to help test a platform for hosting web applications written in an ML-like language? http://www.graftid.com/
Mr_Awesome has quit [Ping timeout: 276 seconds]
Mr_Awesome has joined #ocaml
yziquel has joined #ocaml
sgnb``` has joined #ocaml
sgnb`` has quit [Remote host closed the connection]
<flux> hey it's got colors and everything
<flux> so, graftid is like a web application template system?
<Smerdyakov> I often answer "yes" to not-quite-accurate descriptions, but this one is just too limited, so I'll say "no." :P
<Smerdyakov> (To me, "template" implies Turing-incomplete.)
<flux> well, m4 is sort of a template system, and I think it's turing complete..
<Smerdyakov> Then perhaps you can explain what you mean by "template system."
<flux> it is a system intended to be used for creating output by giving it a "template" and a few parameters that the template requires
<Smerdyakov> I think this doesn't describe Graftid accurately, because Graftid designers are more than just strings with holes in them.
<Smerdyakov> (And the latter pretty universally describes "web template systems.")
<flux> the demo templates don't quite show off those non-string holes yet, though?
<flux> argh, let's say demo system generator examples :)
<Smerdyakov> In a sense they do.
<Smerdyakov> While these designers could be encoded as strings-with-holes, they actually work via a more general code generation system.
<Smerdyakov> Also, the whole building-a-nice-GUI-to-elicit-values-for-the-holes aspect is absent from web template systems, as I understand them.
lokydor has quit [Ping timeout: 240 seconds]
<yziquel> anyone with asm knowledge?
<yziquel> that could tell me how to send a constant into a fixed adress in memory?
middayc has joined #ocaml
<Smerdyakov> yziquel, obviously depends on the instruction set, and you should ask in #asm or some place like that.
Narrenschiff has joined #ocaml
Drk-Sd has joined #ocaml
alinbsp has joined #ocaml
orbitz has joined #ocaml
<yziquel> yes! custom kernel code booting! i was wondering: would it be interesting to have code run in ring 0 in long mode, and do ocaml minor heap garbage collection by raising paging errors and catching them? or in protected mode by catching memory access violations?
<derdon> I want the day to have 48 hours.
<derdon> bye, guys
derdon has quit [Quit: derdon]
avsm has quit [Quit: Leaving.]
Smerdyakov has quit [Quit: Leaving]
orbitz has quit [Ping timeout: 252 seconds]
ulfdoz has quit [Ping timeout: 246 seconds]
orbitz has joined #ocaml
CcSsNET has joined #ocaml
Narrenschiff has quit [Quit: Narrenschiff]
[df]_ has quit [Ping timeout: 252 seconds]
[df]_ has joined #ocaml
det has quit [Remote host closed the connection]
det has joined #ocaml
Amorphous has quit [Ping timeout: 272 seconds]
boscop has quit [Read error: Connection reset by peer]
ikaros has quit [Quit: Leave the magic to Houdini]
pad has joined #ocaml
Amorphous has joined #ocaml
Anarchos has quit [Quit: Vision[0.9.7-H-090423]: i've been blurred!]
hto has joined #ocaml