00:30
rogo has quit ["Leaving."]
00:39
rogo has joined #ocaml
00:57
psnively has quit []
01:03
bzzbzz has joined #ocaml
01:39
jeddhaberstro has quit []
01:41
landonf has quit []
01:41
bluestorm has joined #ocaml
01:42
bluestorm has quit [Client Quit]
01:43
bluestorm has joined #ocaml
01:44
bzzbzz has quit ["leaving"]
01:44
bluestorm has quit [Client Quit]
01:45
bluestorm has joined #ocaml
01:45
<
bluestorm >
sorry for the join/quit noise
02:07
jeddhaberstro has joined #ocaml
02:27
Proteus__ has joined #ocaml
02:29
netx has joined #ocaml
02:33
Associat0r has quit []
02:43
Proteus_ has quit [Read error: 113 (No route to host)]
03:44
bluestorm has quit [Read error: 110 (Connection timed out)]
03:50
jeddhaberstro has quit []
03:56
TaXules has quit [Remote closed the connection]
03:56
TaXules has joined #ocaml
04:46
<
palomer >
flux, awake yet?
05:28
seafood has joined #ocaml
05:29
seafood has quit [Read error: 104 (Connection reset by peer)]
05:30
seafood has joined #ocaml
05:36
palomer has quit [Remote closed the connection]
05:37
Proteus_ has joined #ocaml
05:53
Proteus__ has quit [Read error: 113 (No route to host)]
06:14
seafood has quit [Read error: 110 (Connection timed out)]
06:49
filp has joined #ocaml
06:59
jonafan_ has joined #ocaml
07:05
jonafan has quit [Read error: 60 (Operation timed out)]
07:16
Linktim has joined #ocaml
07:40
Amorphous has quit [Read error: 110 (Connection timed out)]
07:41
filp has quit [Remote closed the connection]
07:42
Amorphous has joined #ocaml
07:50
Jedai has quit [Read error: 110 (Connection timed out)]
08:00
coucou747 has joined #ocaml
08:00
mikezackles has quit [Read error: 110 (Connection timed out)]
08:05
Linktim_ has joined #ocaml
08:11
Linktim_ has quit ["Quitte"]
08:11
Linktim_ has joined #ocaml
08:16
Linktim has quit [Read error: 113 (No route to host)]
08:17
det has quit [Remote closed the connection]
08:21
GustNG has joined #ocaml
08:23
asmanur has joined #ocaml
08:37
coucou747 has quit [Remote closed the connection]
08:39
coucou747 has joined #ocaml
08:55
gaja has quit ["leaving"]
09:34
Yoric[DT] has joined #ocaml
10:00
guillem has joined #ocaml
10:26
ygrek has joined #ocaml
10:46
jlouis has joined #ocaml
10:57
seafood has joined #ocaml
11:21
knucker has joined #ocaml
11:23
<
knucker >
Hello, I have a file "allReader.ml" inside it I am defining "module AllReader : sig .. end = ...."
11:24
Linktim_ has quit ["Quitte"]
11:24
<
knucker >
What I end up with, is a nested module.... AllReader.AllReader ... how do I avoid this?
11:25
<
mfp >
knucker: place the signature in allReader.mli
11:26
<
mfp >
and remove module AllReader .... from the .ml
11:27
<
mfp >
knucker: e.g., a.ml: let foo = (+) a.mli: val foo : int -> int -> int
11:28
<
mfp >
btw. ocamlc -i can generate the interface for you (remove values and make types abstract as needed)
11:33
Linktim has joined #ocaml
11:34
szell` has joined #ocaml
11:35
<
knucker >
makes sense, thanks
11:36
seafood has quit []
11:36
<
knucker >
next step is that I want all my *Reader.ml modules to share the same Reader signature
11:36
<
knucker >
How do I do that... I've see this done through module AllReader : Reader = ... am I not going to end up with the same nesting issue?
11:38
szell has quit [Read error: 110 (Connection timed out)]
11:39
RobertFischer has joined #ocaml
11:40
RobertFischer has quit [Client Quit]
11:45
knucker has quit ["This computer has gone to sleep"]
11:53
hkBst has quit ["Konversation terminated!"]
11:56
bluestorm has joined #ocaml
12:22
szell has joined #ocaml
12:37
filp has joined #ocaml
12:38
szell` has quit [Read error: 110 (Connection timed out)]
12:49
filp has quit ["Bye"]
12:52
knucker has joined #ocaml
12:54
<
Smerdyakov >
knucker, use the [include] signature item in your .mli files.
12:58
thelema has quit [Read error: 110 (Connection timed out)]
13:02
* knucker
googles it up
13:02
Linktim has quit ["Quitte"]
13:04
<
Smerdyakov >
knucker, bad move. Use the manual.
13:05
<
knucker >
Yeah, I'm looking at the Ocaml Oreilly PDF
13:06
<
knucker >
Can't find anything...
13:07
Linktim has joined #ocaml
13:07
<
Smerdyakov >
That's not the manual.
13:08
<
knucker >
which is the manual? The PDF is the main reference I have been using
13:33
asmanur has quit [Read error: 60 (Operation timed out)]
13:42
Associat0r has joined #ocaml
13:53
rwmjones has joined #ocaml
13:54
jlouis has quit [Remote closed the connection]
14:10
jlouis has joined #ocaml
14:19
jlouis has quit [Remote closed the connection]
14:27
redocdam has joined #ocaml
14:30
jlouis has joined #ocaml
14:36
<
gildor >
rwmjones: hi
14:50
vixey has joined #ocaml
14:53
knucker has quit ["This computer has gone to sleep"]
14:55
rwmjones has quit [Read error: 113 (No route to host)]
15:01
ulfdoz_ has joined #ocaml
15:01
ulfdoz_ has quit [Client Quit]
15:17
jlouis has quit [Remote closed the connection]
15:18
pango_ has quit [Remote closed the connection]
15:19
<
mfp >
gildor: I've been reading ocaml-autobuild
15:22
<
mfp >
so you have in place essentially: platform + dep checks, cmdline handling, env/configuration caching, interpolation of $(var) in .in files, calling ocamlbuild/others, right?
15:22
<
gildor >
mfp: yes, that is the idea
15:22
jlouis has joined #ocaml
15:23
<
gildor >
mfp: but it is not a "everyone must use my progam" this is a "I build this for my own use hoping other can use it"
15:23
<
mfp >
atm. you have #directory and #use; I suppose it'd be installed eventually?
15:23
<
gildor >
mfp: FYI the program is one week old ;-)
15:23
<
gildor >
#directory and #use are only here because I am in dev stage
15:24
<
gildor >
i think at the end there will a #ocaml-autobuil ocamlbuild
15:24
<
mfp >
well, the goal would be "everyone want to use this because it makes it easier to develop, release, obtain, install, etc." :-)
15:24
<
gildor >
that will transform in #directory + #use for dev. env
15:25
<
gildor >
and expand to the full source for distribution
15:25
<
mfp >
I'm not sure I get this
15:25
<
gildor >
indeed the goal is "I hope everyone want to use this because"
15:25
<
gildor >
mfp: this is not yet implemented
15:25
<
mfp >
you mean releasing w/ a new toplevel directive in the build.ml?
15:26
<
gildor >
but my idea is to have build.ab (autobuild)
15:26
<
gildor >
that transform into build.ml
15:26
<
gildor >
depending if you are a dev. or a user
15:26
pango_ has joined #ocaml
15:26
<
mfp >
I'd like to have something as simple as foo.cabal
15:26
<
mfp >
so the info can be extracted without running anything
15:27
<
mfp >
the equiv to build.ml is a 3-line Setup.lhs
15:28
<
gildor >
i think it is a good idea
15:28
<
mfp >
all the info is essentially in a very simple file name: foo version: 0.0 ....
15:28
<
gildor >
and you can parse it quite easily
15:28
<
mfp >
want me to write the parser?
15:28
<
gildor >
if you want
15:29
<
mfp >
no external dependencies, right?
15:29
<
gildor >
But you should think "minimal external dependencies"
15:29
<
gildor >
you can use everything in ocaml (Str... Unix but only things that are OK on windows)
15:29
<
gildor >
+ fileutils
15:30
<
gildor >
the idea is to be able to copy and paste the parser directly in build.ml (ie #use)
15:31
<
mfp >
what about creating a sample project so we have something to run this against?
15:31
<
mfp >
say, something like cryptokit, which combines stubs and OCaml code
15:31
<
mfp >
(it's also the example used in ocamlbuild's wiki)
15:31
<
gildor >
i prefer a real "sample" project
15:32
<
gildor >
we don't care if there is something in the project
15:32
<
gildor >
and i don't want license problem by copying other project
15:32
<
mfp >
just to make sure it's doing the right thing
15:32
<
mfp >
OK, dummy code then
15:32
<
gildor >
use a directory name "test" at the root of the project
15:33
<
gildor >
i use oUnit most of the time to do unitary test
15:33
<
mfp >
something with trivial stubs and an executable, so we can darcs setpref the tests
15:36
<
mfp >
what about the "special files" like META, .clib, _tags, .itarget, etc.?
15:36
<
mfp >
they could be generated automatically
15:36
<
mfp >
unless already present, in which case they are preserved
15:37
<
gildor >
yes, but make this an "option"
15:37
<
mfp >
this makes simple things easy (nothing to do besides build.ab), and complex things possible (can hand-write them)
15:38
<
gildor >
you should separate the parsing and the generation of this file
15:38
<
mfp >
for instance with build-type: [simple | custom-ocamlbuild | ...] in build.ab
15:38
<
mfp >
yes, since the parser will be reused when the time comes to extract the info (for the DB, etc.)
15:39
<
gildor >
and you should avoid recreating a file that will have the same conten
15:39
<
gildor >
which will trigger a rebuild in certain case
15:41
<
mfp >
any prefs regarding ocamlyacc vs. ad-hoc String processing for the parser?
15:41
<
gildor >
i do prefer ad-hoc string processing (sorry)
15:41
<
mfp >
the former is "more correct" but takes more work and is messier installation-wise, the later is simpler
15:41
<
gildor >
you can use Str
15:41
<
mfp >
me too in this case actually :)
15:42
<
gildor >
and Genlex
15:46
<
gildor >
mfp: if you are interested in cooperation, i will setup a project on forge.ocamlcore.org
15:47
<
mfp >
I was about to suggest it
15:47
<
gildor >
keep your changes into your local copy of darcs
15:47
<
gildor >
I will setup the project after my holiday (15th August)
15:48
<
mfp >
does it require manual intervention? iow. cannot I create the project as a normal forge user (once I register)?
15:49
<
gildor >
it needs admin intervention for creating the darcs repository
15:49
<
gildor >
but we are not in a hurry
15:50
<
gildor >
you can record everything you want inside your darcs local repository and then send it to the project we will setup
15:50
<
Yoric[DT] >
For a darcs repository ?
15:50
<
Yoric[DT] >
I thought ssl + http was sufficient ?
15:50
<
mfp >
ssl? do you mean SSH?
15:50
<
gildor >
Yoric[DT]: yes, but and admin must create the directory hosting the project with the good right
15:51
<
gildor >
mkdir darcsroot/XXX + chmod g+w darcsroot/XXX + chown scm-forge ...
15:52
* mfp
has always darcs pushed through SSH --- don't remember if it can also do HTTP PUT, but it'd need extra setup in the server anyway
15:58
<
Yoric[DT] >
mfp: yes, I meant ssh.
15:58
<
Yoric[DT] >
gildor: ok.
16:09
jeddhaberstro has joined #ocaml
16:39
marmotine has joined #ocaml
16:40
tomh- has joined #ocaml
16:46
tomh- is now known as tomh
16:47
Kopophex has joined #ocaml
17:05
Watza has joined #ocaml
17:05
<
Watza >
donc comme je vous le disais il y a quelque temps
17:05
<
Watza >
j'ai des insomnies de merde !!!
17:05
Watza has left #ocaml []
17:30
<
bluestorm >
gildor: do project owner get notified when someone submit a new patch ?
17:32
<
mfp >
AFAIK darcs doesn't have any hook for push
17:32
<
mfp >
git does, however
17:32
<
bluestorm >
mfp: i used the forge "tracker" system
17:34
<
mfp >
that might be user-configurable (I remember receiving notices w/ some forges and not getting anything w/ some other)
17:42
asmanur has joined #ocaml
18:21
<
gildor >
bluestorm: I think you can configure it
18:21
<
gildor >
but i have not try
18:22
<
gildor >
when people submit bug to site admin, i get notified
18:22
<
gildor >
i think it is the same principle
18:46
mikezackles has joined #ocaml
18:51
asmanur has quit [Read error: 110 (Connection timed out)]
19:17
Linktim has quit [Read error: 110 (Connection timed out)]
19:19
palomer has joined #ocaml
19:20
<
palomer >
flux, you awake?
19:20
Linktim has joined #ocaml
20:26
<
flux >
palomer, maybe you could try stracing the binary, if you're on linux
20:26
<
flux >
I'm guessing it blocks somewhere
20:27
<
flux >
alternative to stracing is to add debug trace messages
20:27
<
flux >
also the channel isn't in nonblocking mode
20:28
<
flux >
moreover you're using open_process_full, but you're never providing it any input. shoudln't matter with ls, though.
20:29
<
flux >
also, you're not consuming err_c, but that shoudln't matter in this case either
20:29
<
flux >
damn, typoing that should-word alot
20:30
<
flux >
does it work without the add_watch-thingy?
20:54
ygrek has quit [Remote closed the connection]
21:01
mattam has quit ["Lost terminal"]
21:19
mattam has joined #ocaml
21:37
seafood has joined #ocaml
21:38
Linktim has quit ["Quitte"]
21:43
seafood has quit [Read error: 54 (Connection reset by peer)]
21:44
seafood has joined #ocaml
21:45
seafood has quit [Read error: 104 (Connection reset by peer)]
21:45
seafood has joined #ocaml
21:46
seafood has quit [Read error: 104 (Connection reset by peer)]
21:47
seafood has joined #ocaml
21:48
seafood has quit [Read error: 104 (Connection reset by peer)]
21:49
seafood has joined #ocaml
21:54
seafood has quit [Read error: 104 (Connection reset by peer)]
21:54
seafood has joined #ocaml
22:05
seafood has quit [Connection reset by peer]
22:06
seafood has joined #ocaml
22:16
seafood has quit [Read error: 54 (Connection reset by peer)]
22:16
seafood has joined #ocaml
22:29
thelema has joined #ocaml
22:37
<
palomer >
flux, still around??
22:37
<
palomer >
flux, "running" is never printed
22:41
seafood has quit [Read error: 110 (Connection timed out)]
23:04
* palomer
hates unix.read!
23:14
jlouis has quit ["Leaving"]
23:28
rogo has quit [Read error: 110 (Connection timed out)]
23:35
coucou747 has quit ["bye ca veut dire tchao en anglais"]
23:47
Yoric[DT] has quit ["Ex-Chat"]
23:49
seafood has joined #ocaml
23:50
seafood has quit [Read error: 104 (Connection reset by peer)]
23:50
seafood_ has joined #ocaml
23:50
seafood_ has quit [Client Quit]