kakadu_ has quit [Remote host closed the connection]
sapristi has quit [Remote host closed the connection]
sapristi has joined #ocaml
Denommus has joined #ocaml
sapristi has quit [Ping timeout: 268 seconds]
abhir00p has joined #ocaml
<abhir00p>
I am trying to install opam on my OSX Yosemite
<abhir00p>
on brew install opam
<abhir00p>
it installs ocaml, ocamlbuild, camlp4, boost
<abhir00p>
but after that it hangs
<abhir00p>
the cpu is at full consumption
<abhir00p>
its trying to build something
<abhir00p>
one of these : sphinx-doc, cmake, re2c, gringo, clasp, aspcud
<p1n34ppl3>
not sure why any of those would be a dependency, but maybe someone else does :p
<abhir00p>
okay I tried `brew install opam --without-aspcud` and it worked
<abhir00p>
the docs mention thats for OSX Mountain Lion or lower
<jmiven>
aspcud is the dependency solver used by opam. Gringo and clasp are used by aspcud
<p1n34ppl3>
oh
<jmiven>
abhir00p: are you sure that the build is "hanged"? aspcud, clasp and gringo are C++ dependencies and they can be quite long to build
kalio has joined #ocaml
<abhir00p>
No I don't think it actually hung it was perhaps taking a long time to build but CPU was at full consumption for nearly 15 mins and no activity
<abhir00p>
I eventually downloaded, I hope the abscence of aspcud won't be an issue.
<jmiven>
abhir00p: in my experience opam doesn't work very well without it, you might experience some problems
<abhir00p>
Ah I think the culprit is boost
<abhir00p>
boost is one of the dependency of aspcud
<abhir00p>
yes there is a build script called b2 which hogs all the memory
<dhtns>
I installed opam 1.2.2 from the official repositories (arch linux)
<p1n34ppl3>
what i did was install ocaml from the repos, install opam manually to ~/.local and opam switch 4.06.0, then put the init stuff in my bash profile
<p1n34ppl3>
seems like you are missing acpusd which is the solver opam uses
<dhtns>
p1n34ppl3: that command exists though (it was pulled in as a dependency to opam)
<dhtns>
Running with --use-internal-solver fixes that particular error, but I got different errors later on so I decided to try and do everything correctly from the start :)
<p1n34ppl3>
let me get on pc lol
<dhtns>
I read in some issue that 'acpusd' needs gringo and clasp as dependencies, but they weren't automatically pulled in
<p1n34ppl3>
seems like i only have the opam binary itself under local
<p1n34ppl3>
and i dont seem to have any of the external solvers installed, and im not getting any errors
<p1n34ppl3>
using arch too
<p1n34ppl3>
odd, because i forgot what i actually did lol
<dhtns>
p1n34ppl3: Do you have clingo installed?
<dhtns>
community/clingo
<p1n34ppl3>
ohh yes clingo
<p1n34ppl3>
yeah
<dhtns>
Oh hey, seems like it works if I install clingo :)
<p1n34ppl3>
nice :d
<dhtns>
Shouldn't clingo be marked as a dependency to acpusd then, though?
<p1n34ppl3>
i eventually ended up uninstalling the official packages opam and ocaml so using only a local version of opam and opam switch 4.06.0 which installs stuff under ~/.opam
<dhtns>
p1n34ppl3: So essentially, everything ocaml-related lives under .opam?
<dhtns>
(corollary: rm -rf .opam nukes your entire installation?)
<p1n34ppl3>
yes, and opam in ~/.local/bin
<p1n34ppl3>
yeah
<dhtns>
okay, I like that :D
<p1n34ppl3>
and i had to do it because using the system switch wasnt working properly
<dhtns>
side-note: why are there so many possible compiler versions in 'opam switch'?
<p1n34ppl3>
so you can have more versions of the compiler, i mainly use it to test if my code runs on older versions
<p1n34ppl3>
but can remove them
<Drup>
Well, there are all the OCaml versions, plus the compilers with some options enabled, plus some various forks
<p1n34ppl3>
i mean, if installed
<dhtns>
i think it's really useful to have that feature, but it's kinda weird that you have to explicitly 'select' a version to install before you can compile stuff
<dhtns>
I'd expect that you'd just get a stable version installed by default
<dhtns>
as a beginner it feels like I need to make an informed decision as to which compiler I'm going to be using
jao has joined #ocaml
zolk3ri has joined #ocaml
<zolk3ri>
dhtns: Do you have ". /home/zolk3ri/.opam/opam-init/init.sh > /dev/null 2> /dev/null || true" in ~/.bash_profile or ~/.bashrc?
<zolk3ri>
Err, ~/ :P
<dhtns>
zolk3ri: Yes, modulo my username :p
<zolk3ri>
You don't have to explicitly do an opam switch if you intend to use the last version (4.06.0) and already done `opam switch 4.06.0' and have that init.sh stuff in any of these files. Be aware that ~/.bash_profile gets executed only once, so you would have to actually log in and log out, running another bash instance won't work.
<Drup>
you certainly don't have to log ing/log out
<zolk3ri>
You don't, if you use ~/.bashrc or if you source .bash_profile in your .bashrc.
<Drup>
First, just put it in your .bashrc instead. Second, You can just run `eval $(opam config env)` after switching
<dhtns>
zolk3ri: If I'm understanding you correctly, I only need to do an explicit `opam switch` once (to install the latest version)?
<dhtns>
I have the init.sh stuff in my shell config
<dhtns>
(I tested that the environment variables were actually set correctly in a new shell, and that seemed to be the case)
<zolk3ri>
dhtns: Yes.
<dhtns>
Okay, then I think I'm set
<zolk3ri>
If you type "opam switch", you can see which one is being used.
<dhtns>
that tells me I'm on the latest version of the compiler
<dhtns>
(that undefined symbol is in fact defined in the so)
Denommus has quit [Ping timeout: 255 seconds]
<Drup>
Z3 was not "removed". It was simply never added :3
<Drup>
or, well, added only very briefly, but the merge can be considered a short-lived mistake
<dhtns>
After installation of Z3, opam prompted that I should add "$(ocamlfind printconf destdir)/stublibs" to LD_LIBRARY_PATH but that doesn't change anything
<Drup>
Anyway, this should clearly be revisted, I did the package a long time ago, when the Z3 situation was quite messy
<Drup>
dhtns: I'm not sure why you need to make a commit to add an opam repo, that's not supposed to be the case
<dhtns>
(Drup: note that I didn't explicitly need to make one, it happens under the hood, I just noticed it because it asks me to sign the commit)
<Drup>
yeah, but even that shouldn't happen, it's .. strange
malina has joined #ocaml
<Drup>
dhtns: that's when you do "opam repo add" ?
<dhtns>
Yup
<Drup>
where does it commit ?
<dhtns>
I honestly have no idea, I didn't want to dig too deep
<dhtns>
I just strace'd the opam repository add command
rwmjones has quit [Ping timeout: 255 seconds]
<dhtns>
I gave up when I saw it made a harmless commit
<Drup>
Could you make a bug report to opam ? That's shouldn't happen
<dhtns>
Drup: I'm gonna get back to it, but I'm hungry so I'm getting food now :)
<Drup>
fair enough :p
rwmjones has joined #ocaml
freusque has joined #ocaml
Denommus has joined #ocaml
<companion_cube>
Drup: so, is there any obstacle right now to repackaging z3 for opam?
<companion_cube>
and your wrapper in another package that depends on it?
<Drup>
There are always obstacles :D
<Drup>
but it might be worth it to give it another try, at least
<Drup>
the main question is: do we recompile Z3 or not
<companion_cube>
what are the obstacles?
<companion_cube>
there could be a conf-z3 package?
<companion_cube>
that relies on the .so
<companion_cube>
ah wait, the bindings would not be there, would they…
<Drup>
:]
<Drup>
and compiling just the bindings might involve some not-so-pleasant buildsystem hacks
<companion_cube>
yeah yeah
<companion_cube>
the clean way, imho, would be for modern OCaml bindings to directly use Z3's C API
<companion_cube>
but that's a lot of work
<companion_cube>
so, maybe the simplest way right now is to build z3 ourselves
<Drup>
that's a lot of work, yes
<Drup>
the current binding's API is horrible, but it deals with memory properly, which is ... non trivial
<companion_cube>
:D
<companion_cube>
your own wrapper builds on it, right?
<Drup>
yes, that was the point
<Drup>
(although it could also be improved)
<Drup>
(I tried a bit to hard to avoid float/integer casts)
<companion_cube>
well I'm interested in contributing to this packaging and to your binding, in general
<companion_cube>
if you want to make a separate repo
<companion_cube>
well… it takes 12 min to compile on my laptop, tho, so I'm not sure travis will be pleased
<Drup>
when I worked on this, with my old laptop, it would be more like 30/35 mins
<Drup>
it was not enjoyable
sgnb` has quit [Ping timeout: 268 seconds]
<Drup>
Also, Z3's incremental compilation is not very robust
sgnb` has joined #ocaml
ziyourenxiang has quit [Ping timeout: 260 seconds]
yomimono has quit [Quit: Leaving]
andreas___ has quit [Quit: Connection closed for inactivity]
<companion_cube>
:D
<companion_cube>
well, maybe that's why they're moving towards cmake
<companion_cube>
also, C++11 ♥
<companion_cube>
(sorry, not related, but good nonetheless)
p1n34ppl3 has joined #ocaml
slash^ has joined #ocaml
sapristi has quit [Remote host closed the connection]
<p1n34ppl3>
_2
<companion_cube>
_3
kakadu has joined #ocaml
argent_smith has quit [Quit: Leaving.]
jao has joined #ocaml
cbot has joined #ocaml
<reynir>
_4
<Jfault>
_5
AlexRussia has joined #ocaml
<zolk3ri>
_6
kerrhau has joined #ocaml
kerrhau has quit [Changing host]
kerrhau has joined #ocaml
andreas___ has joined #ocaml
jnavila has joined #ocaml
dtornabene has quit [Remote host closed the connection]
<Jfault>
anyone have ideas about my opam problem?
<Jfault>
ugh, I really need this done soon
<Jfault>
might have to self install :(
slash^ has quit [Read error: Connection reset by peer]
enterprisey has quit [Remote host closed the connection]
Taneb has joined #ocaml
<Taneb>
Hi! I'm a Haskell programmer looking to dabble with OCaml... is there any tutorial/guide aimed at people already experienced with strongly typed functional programming?
johnv has joined #ocaml
<dhtns>
Drup: I looked at the weird behaviour of "opam repo add" again just now
<dhtns>
I found this: [pid 16745] execve("/usr/bin/git", ["git", "commit", "--allow-empty", "-m", "opam-git-init"], 0x565467475d10 /* 75 vars */) = 0
<dhtns>
however I don't have the slightest clue what piece of code could be calling that command
_andre has quit [Quit: leaving]
<dhtns>
I searched through the source code of opam to no avail
johnv has left #ocaml [#ocaml]
<zolk3ri>
Jfault: Perhaps clone the git repo and compile it yourself after you removed opam via pacman, then do "opam switch 4.06.0" and follow the instructions.
<Jfault>
Taneb: take a look at the docs
<Jfault>
you need to get the syntax and stuff down
<Jfault>
zolk3ri: installing opam from git worked without any other changes
<Jfault>
so must have been the version of opam in the arch repo?
<Drup>
dhtns: just report it
<Taneb>
Jfault, thanks for the suggestion
<Jfault>
tbh I like coding in haskell more (unpopular opinion around here), but I need ocaml for this one thing I'm working on
<Taneb>
Jfault, that's mostly the same with me, :D
ygrek has joined #ocaml
argent_smith has joined #ocaml
zv has quit [Ping timeout: 248 seconds]
zv has joined #ocaml
Denommus has quit [Remote host closed the connection]
<Jfault>
getting merlin set up right now, actually
<Jfault>
how great is it :P
pmetzger has joined #ocaml
<whoman>
much =)
snhmib has joined #ocaml
<whoman>
just dont forget .merlin and the PKG line and its flawless
shinnya has joined #ocaml
picolino has joined #ocaml
TarVanimelde has joined #ocaml
pmetzger has quit [Read error: Connection reset by peer]
mk9_ has joined #ocaml
mk9_ has quit [Read error: Connection reset by peer]
mk9 has joined #ocaml
sh0t has quit [Remote host closed the connection]
sh0t has joined #ocaml
whoman has quit [Read error: Connection reset by peer]
mk9_ has joined #ocaml
tane has joined #ocaml
mk9 has quit [Ping timeout: 248 seconds]
TarVanimelde has quit [Quit: TarVanimelde]
TarVanimelde has joined #ocaml
TarVanimelde has quit [Client Quit]
TarVanimelde has joined #ocaml
TarVanimelde has quit [Client Quit]
TarVanimelde has joined #ocaml
mk9_ has quit [Quit: mk9_]
TarVanimelde has quit [Client Quit]
TarVanimelde has joined #ocaml
TarVanimelde has quit [Client Quit]
TarVanimelde has joined #ocaml
TarVanimelde has quit [Client Quit]
zmt00 has quit [Quit: Leaving]
pmetzger has joined #ocaml
jnavila has quit [Remote host closed the connection]
sh0t has quit [Remote host closed the connection]
sh0t has joined #ocaml
argent_smith has quit [Quit: Leaving.]
AltGr has left #ocaml [#ocaml]
nojb has quit [Ping timeout: 248 seconds]
kerrhau has quit [Ping timeout: 248 seconds]
kerrhau has joined #ocaml
kerrhau has quit [Changing host]
kerrhau has joined #ocaml
zolk3ri has quit [Quit: leaving]
AlexRussia has quit [Ping timeout: 240 seconds]
SomeDamnBody has joined #ocaml
<SomeDamnBody>
Has anybody ever had to deal with singularity before?
<companion_cube>
every time I divide by 0
nullifidian has quit [Remote host closed the connection]
snhmib has quit [Ping timeout: 258 seconds]
nullifidian has joined #ocaml
nojb has joined #ocaml
raphaelss has joined #ocaml
AlexRussia has joined #ocaml
pierpa has joined #ocaml
mk9 has joined #ocaml
Jesin has quit [Quit: Leaving]
M__ has joined #ocaml
<M__>
when installing tls with opam, I get the following error:
<M__>
#=== ERROR while installing asn1-combinators.0.1.3 ============================#
<M__>
# File "src/asn_prim.ml", line 120, characters 60-409: # Error: Signature mismatch: # ... # Values do not match: # val random : ?size:int -> unit -> bytes # is not included in # val random : ?size:int -> unit -> t # File "src/asn_prim.ml", line 13, characters 2-37: Expected declaration # File "src/asn_prim.ml", line 128, characters 6-12: Actual declaration
<M__>
Any idea?
whoman has joined #ocaml
ygrek has quit [Ping timeout: 268 seconds]
<companion_cube>
if you use 4.06, this package might not be up to date yet
<companion_cube>
try `opam switch 4.05.0` maybe
<companion_cube>
(I mean OCaml 4.06; it definitely separates `string` and `bytes`)
<M__>
I think you are right
<M__>
Thanks
tane has quit [Quit: Leaving]
mk9 has quit [Quit: mk9]
navaja has joined #ocaml
sepp2k has quit [Quit: Leaving.]
andreas___ has quit [Quit: Connection closed for inactivity]
M__ has quit [Ping timeout: 260 seconds]
<whoman>
there could/should be a way to use merlin's data in toplevel? it seems silly to eval/compile everything twice.
<companion_cube>
the toplevel evaluates only once, what do you mean?
<whoman>
well, merlin does a bunch of parsing and stuff on emacs buffer save
<whoman>
but im then evaluating my changes into the toplevel
Anarchos has quit [Quit: Vision[0.10.2]: i've been blurred!]
<whoman>
keeping the two in sync sort of feeling, if that makes sense
<whoman>
it might be more of an #emacs thing than an #ocaml thing, i cant tell
kerrhau has quit [Quit: bye!]
navaja has quit [Ping timeout: 240 seconds]
Algebr has joined #ocaml
<companion_cube>
hmmm
<companion_cube>
the toplevel doesn't know about merlin anyway
kerrhau has joined #ocaml
spew has quit [Quit: Connection closed for inactivity]