<Intensity>
Hi. I'm trying to update libdose3-ocaml on Debian 7 so that I can backport opam, so that I can more easily install google-drive-ocamlfuse. I'm having trouble with ocamlfind: Package `re.pcre' not found.
Submarine has quit [Remote host closed the connection]
<Intensity>
whitequark: Thanks for the reference. I'm wondering if there's been a rename (from re.pcre |-> pcre) lately, and what the statement "module Pcre = Re_pcre" in common/cudfSolver.ml might really mean, given that Re_pcre is only referenced dose3-3.2.2, and not otherwise defined/declared.
WraithM has quit [Ping timeout: 252 seconds]
<whitequark>
no, there was not a rename
<whitequark>
the package is called re.pcre, the library is called Re_pcre
<whitequark>
the statement you refer to creates a module alias
<whitequark>
i.e. after that statement, it is possible to refer to Re_pcre as Pcre
<Intensity>
whitequark: If there was not a rename, do you know why I got ocamlfind: Package `re.pcre' not found, when ocamlfind ocamldep was run, or why I may have gotten "Error: Unbound module Re_pcre" when compiling common/cudfSolver.ml?
<Intensity>
Then do you know why "ocamlfind ocamldep -package pcre" might return success? query re indicates re is not installed. So is it possible that I have pcre but not re.pcre, and I somehow need the latter?
<whitequark>
this is exactly what happened
<whitequark>
pcre and re.pcre are completely unrelated
<whitequark>
pcre is a binding for libpcre; re (and re.pcre) are purely in OCaml
<whitequark>
I admit that it is confusing.
<Intensity>
whitequark: Okay, so then does that mean they can coexist? I'm wondering if there is a Debian package for the ocaml "re". Maybe in testing or in wheezy.
<whitequark>
try installing the package libre-ocaml-dev
<whitequark>
yes, they can coexist
<whitequark>
re is in jessie/sid
<Intensity>
Cool, found that. I'm getting that now.
rgrinberg has quit [Quit: Leaving.]
<Intensity>
Hmm. Seems that I get "Error: Unbound module Graphml" even though libocamlgraph-ocaml-dev-1.8.2-2 is installed.
<Intensity>
Maybe I need 1.8.5. I'll try that.
<whitequark>
you may need to add the corresponding -package
WraithM has joined #ocaml
<whitequark>
in this case ... seems to be just -package ocamlgraph
<Intensity>
Yes, "-package ocamlgraph" was included in the command line that failed.
cdidd_ has joined #ocaml
cdidd has quit [Ping timeout: 260 seconds]
A1977494 has joined #ocaml
A1977494 has quit [Quit: Leaving.]
martintrojer has quit [Max SendQ exceeded]
martintrojer has joined #ocaml
Hannibal_Smith has joined #ocaml
martintrojer has quit [Read error: Connection reset by peer]
<Drup>
we will have fibers bolted in the compiler for concurrency, and domains for big grain parallelism, with auto balancing of fibers across domains.
<whitequark>
I see, thanks
<whitequark>
this is an odd decision
<whitequark>
but interesting
<Drup>
native code, lazy, and various stuff are broken, but bytecode is working :p
ygrek has joined #ocaml
samrat has joined #ocaml
burgobianco has quit [Remote host closed the connection]
burgobianco has joined #ocaml
WraithM has quit [Ping timeout: 252 seconds]
shinnya has joined #ocaml
samrat has quit [Quit: Computer has gone to sleep.]
ygrek has quit [Ping timeout: 264 seconds]
MercurialAlchemi has quit [Ping timeout: 268 seconds]
hhugo has joined #ocaml
tobiasBora has joined #ocaml
<tobiasBora>
Hello,
<tobiasBora>
I've a little problem with a fresh install of opam
<tobiasBora>
I have a project which is based on oasis, but when I do ./configure I've the error E: Cannot find findlib package unix
<tobiasBora>
E: Cannot find findlib package str
<tobiasBora>
And same thing for lot's of packages that have already been installed by opam.
<Drup>
create a dummy one only to report issues and stop whining ? :p
<adrien>
ah
<whitequark>
adrien: does your religion forbid you to use github?
<adrien>
create a dummy account so I can whine so I stop whining :P
<adrien>
crap ToS
<Drup>
I have one on mantis and the ocaml forge, even if they are crappy tool written in languages I dislike. I'm sure you can overcome your dislike for github's license.
<whitequark>
uhhh, what exactly is broken in GitHub's TOS?
zpe has joined #ocaml
<whitequark>
they have probably the sanest ToS out of ... pretty much all ones I have read
ustunozgur has joined #ocaml
<whitequark>
I think they had some weird clause back in 2008, which is why I didn't use it back then
<adrien>
but overall it looks better than it used to
<whitequark>
uhhh, what?
tobiasBora has quit [Remote host closed the connection]
<whitequark>
I can understand F3, but what's your problem with G*?
<adrien>
"sole discretion"
<adrien>
G11 is also clearly subjective
<adrien>
and a call to unlink() fits in it
<adrien>
s/G100/G10/ btw
<whitequark>
sole discretion, of course
<adrien>
california law: I don't feel like doing stuff the california way
<whitequark>
ahem
<whitequark>
so what's the *real* reason you avoid it?
<adrien>
these and I dislike the software UI
dsheets has quit [Ping timeout: 252 seconds]
<whitequark>
so, do you use Google? ;P
<adrien>
no
<adrien>
or extremely rarely
<adrien>
(and most of the time I don't get the answer I want from it anyway)
cdidd_ has joined #ocaml
tobiasBora has joined #ocaml
BitPuffin has joined #ocaml
hhugo has quit [Quit: Leaving.]
cthuluh has quit [Remote host closed the connection]
ollehar has quit [Ping timeout: 264 seconds]
<whitequark>
def`: huh, merlin fails with Error_forward on OUnit2 too
rgrinberg has joined #ocaml
rgrinberg has quit [Client Quit]
oscar_toro has joined #ocaml
rand000 has quit [Ping timeout: 268 seconds]
gperetin has quit [Ping timeout: 260 seconds]
esden has quit [Ping timeout: 260 seconds]
abbe has quit [Ping timeout: 260 seconds]
Leonidas has quit [Ping timeout: 260 seconds]
Leonidas has joined #ocaml
rand000 has joined #ocaml
abbe has joined #ocaml
AltGr has joined #ocaml
gperetin has joined #ocaml
hexo has quit [Ping timeout: 260 seconds]
ygrek has joined #ocaml
hexo has joined #ocaml
esden has joined #ocaml
rgrinberg has joined #ocaml
rgrinberg has quit [Quit: Leaving.]
hhugo has joined #ocaml
ontologiae has joined #ocaml
koderok has joined #ocaml
ustunozgur has quit [Remote host closed the connection]
ygrek has quit [Ping timeout: 245 seconds]
fold has joined #ocaml
samrat has joined #ocaml
Thooms has joined #ocaml
q66 has joined #ocaml
leowzukw has joined #ocaml
hhugo has quit [Quit: Leaving.]
oscar_toro has quit [Quit: oscar_toro]
msaegesser has joined #ocaml
tobiasBora has quit [Ping timeout: 264 seconds]
dsheets has joined #ocaml
darkf has quit [Quit: Leaving]
MercurialAlchemi has joined #ocaml
<whitequark>
def`: nevertheless, merlin makes me *much* more productive
<Drup>
btw whitequark
<Drup>
are you aware on the initiative to make ocaml work with gdb ?
<whitequark>
uh. it works with gdb
<whitequark>
I have used it :)
<Drup>
better :D
<whitequark>
okay. no, I'm not
<whitequark>
it sounds straightforward but extremely tedious
<whitequark>
*I've not
<whitequark>
*I'm not
<whitequark>
ARGH
<Drup>
apparently, it has type reconstruction and breakpoing and stuff
<whitequark>
not aware.
<whitequark>
breakpoints work
<whitequark>
type reconstruction? what do you mean?
<whitequark>
gdb is quite skewed to C/C++
<Drup>
don't know the details
<Drup>
just reporting what anil said, since you were working on your own debuger
<whitequark>
my one is bytecode only and will forever remainso
<ggole>
So it traces pointer paths through the program in order to print, say, constructor names accurately?
<whitequark>
you can't meaningfully turn gdb into an ocaml repl
<ggole>
No, but you can (probably) print shit.
<whitequark>
at most you could make it execute simple expressions, but even that is extremely hard
<whitequark>
(simple as in, pure)
<ggole>
Even floats would be hard
<ggole>
"Oops, need to GC now."
<whitequark>
with an LLVM backend though you could use lldb and JIT OCaml code
<whitequark>
(but I'm not convinced that a native-code REPL is useful enough to warrant that)
<def`>
whitequark: I fixed the Error_forward error that escaped yesterday (error management is more displicined in 4.02_, but I was still doing it <=4.01 way)
<whitequark>
def`: btw can you add src/ocaml to gitignore
<whitequark>
I have this commit dangling on my local copy of merlin2 and it's annoying
<def`>
which commit ?!
<def`>
I will push this change wait
<whitequark>
it's just locally
hhugo has joined #ocaml
<whitequark>
ooooooo
<whitequark>
it now works properly
<whitequark>
well, almost, it is angered by uninterpreted extensions
<def`>
Ok, I pushed both
pyon has quit [Quit: brb]
pyon has joined #ocaml
<whitequark>
great, thanks
lbaan has joined #ocaml
<ggole>
whitequark: what do you use most? Completion and jump-to-def?
<whitequark>
ggole: neither
<ggole>
Oh? Just errors?
<whitequark>
I didn't remember that jump-to-def exists, and completion only half works on this install
<whitequark>
well, I use completion, but the most productivity comes from errors
<whitequark>
actually, no, it's errors and completion for sure. I'm just used to completion, since I used ocp-index for so long
<ggole>
Hmm, I find them a bit annoying and usually have them off.
<whitequark>
I mismatch parens, braces and brackets every damn time
koderok has quit [Ping timeout: 268 seconds]
<ggole>
Probably just what I'm used to.
<whitequark>
and merlin indicates this both quicker and visually more convenient than make
<whitequark>
merlin will work even better for me once def` adds the ppx magic :p
<ggole>
Your editor doesn't do matching?
<whitequark>
it does
<whitequark>
it doesn't complain about mismatched parens though
<whitequark>
and also some mismatches are sensible from editor's POV but not ocaml's parser
<ggole>
Mmm, I see
<ggole>
Interesting how differently people use their tools
<ggole>
Emacs will highlight a mismatched paren, but only under point.
<whitequark>
I wrote a lot of C/C++ (and a bit of Java) without any kind of IDE features whatsoever
<ggole>
(Or if you invoke a matching tool, but that's no easier than simply compiling.)
<whitequark>
so I suppose it's hard to scare me with a little OCaml code
<whitequark>
not to mention Ruby, where you can barely have any kind of IDE at all
<ggole>
Yeah. I dunno how I'd handle having to write a ton of something like Javascript.
<ggole>
Probably by making a large number of mistakes.
<osa1>
I'm trying to install `llvm` from opam but it's failing with `/usr/lib64/gcc/x86_64-suse-linux/4.8/../../../../x86_64-suse-linux/bin/ld: cannot find -lLLVM-3.4`. I have llvm 3.4 installed but I guess it's name is just LLVM instead of LLVM-3.4. Is there a way to tell opam that?
<whitequark>
sigh~
AltGr has left #ocaml [#ocaml]
<whitequark>
osa1: no, there is not
ustunozgur has joined #ocaml
<osa1>
cool. any other ideas how to fix this?
fold has quit [Ping timeout: 255 seconds]
<whitequark>
is your system LLVM only available as libLLVM.so* ?
mhi^ has joined #ocaml
<osa1>
I have libLLVM*.so files for 3.4 in /usr/lib
<whitequark>
so libLLVM.so or libLLVM-3.4.so ?
<osa1>
libLLVM.so
<whitequark>
ok
<def`>
whitequark: I forgot to push jumpnto def
<whitequark>
osa1: this is a bit odd
<whitequark>
the configure script seems to recognize both llvm-config-3.4 and llvm-config
<def`>
whitequark: pushed to sublime-text-merlin repo
<whitequark>
osa1: do you have both of these executables around or only one of them?
<whitequark>
cc jpdeplaix
<osa1>
whitequark: I have only llvm-config
<whitequark>
osa1: ok, can you post the full build log somewhere?
<whitequark>
opam tells you where it is located, at failure
<whitequark>
all recent LLVM, I think since 3.2, has this in the name
<osa1>
so is there hope?
<whitequark>
but before that it used to not have it
<jpdeplaix>
osa1: what's in /usr/lib64/libLLVM*
<jpdeplaix>
?
ontologiae has quit [Ping timeout: 272 seconds]
samrat has quit [Ping timeout: 246 seconds]
Simn has quit [Ping timeout: 268 seconds]
yacks has quit [Read error: Connection reset by peer]
samrat has joined #ocaml
Ptishell has joined #ocaml
Thooms has quit [Quit: WeeChat 0.4.3]
yacks has joined #ocaml
samrat has quit [Quit: Computer has gone to sleep.]
ygrek has joined #ocaml
<whitequark>
function `List ((`String "B")::x::[]) -> (function `Int y -> `Ok | _ -> `Error) x
<whitequark>
why does this fail to typecheck?
hhugo has quit [Quit: Leaving.]
<ggole>
x is being inferred as [< `String of string], I think
<whitequark>
but why?
<whitequark>
if I constrain the outer function argument to Yojson.Safe.json, it works properly
<whitequark>
some bizarre subtyping behavior
<Drup>
whitequark: the second function is assumed to be total
<Drup>
oh, hum, doesn't change the weird behavior
<Drup>
that's strange
jao` has joined #ocaml
jao` has quit [Changing host]
jao` has joined #ocaml
cthuluh has joined #ocaml
osa1 has left #ocaml ["Konversation terminated!"]
leowzukw_ has quit [Quit: leaving]
samrat has joined #ocaml
ontologiae has joined #ocaml
NoNNaN has joined #ocaml
libertas has joined #ocaml
Simn has joined #ocaml
fold has quit [Ping timeout: 252 seconds]
hhugo has joined #ocaml
hhugo has quit [Client Quit]
rand000 has quit [Quit: leaving]
ygrek has quit [Ping timeout: 252 seconds]
tg has quit [Ping timeout: 240 seconds]
tg has joined #ocaml
hhugo has joined #ocaml
mads- has joined #ocaml
<mads->
Can anyone recommend a good resource for learning OCaml? I don't really have any background in functional programming other than a few months of Haskell
<flux>
the multicore with fibers and domains seems pretty interesting.
<flux>
but I think it's not the only ongoing project for multi-core?
<flux>
infact, is that the third one?-o
<Drup>
yes indeed
<Drup>
it's the only *ongoing* one
<Drup>
the others one are dead
<flux>
so new projects pop up while the old ones get abandoned ;)
<Drup>
I really hope it's the last one.
<flux>
well, doesn't need to be the last one but at least it should reach production quality
<Drup>
more than production quality, I want it to be merged >_>
<whitequark>
is it the one with third heap generation?
<whitequark>
minor - major - shared ?
<Drup>
no
<Drup>
at least, not anymore
<Drup>
the major is shared
<whitequark>
O_o
<whitequark>
that got abandoned?
<Drup>
I think it was the initial plan, but they changed it
<ggole>
Soon we'll have as many multicore projects as build systems.
<whitequark>
tbh I liked that more than the weird fiber thing
<whitequark>
I mean, we will now have: lwt threads, normal threads, fibers and multicore domains
<flux>
yeah, I think the some folks don't really get it, we don't want to run a different multicore system per each core :P
<whitequark>
what?
<Drup>
whitequark: "that" = ?
<flux>
;-)
<whitequark>
Drup: that = third generation
<Drup>
ah, right
<Drup>
don't see the link to the fact that you have a notion for concurency and one for //
<Drup>
(which is why fibers ≠ domains)
<whitequark>
why do you even *need* fibers?
<flux>
whitequark, compared to real operating system threads you mean?
<flux>
whitequark, they are super fast for contect switching of course, not to mention the GC aspects..
<whitequark>
flux: we already have lwt though
<Drup>
whitequark: go watch the video already, instead of making us play the repeating game, with partial informations and that takes as much time for everyone.
<whitequark>
Drup: I'll rather just shut up
<flux>
I suppose you could use lwt instead of fibers. I'm not going to say it's more convenient, though.
<flux>
I think fibers don't need to be co-operative like lwt does
<Drup>
(fibers are not monadic)
<flux>
process a big data structure without inserting yielding points inside it -> boom, core is blocked
<Drup>
(and don't suffer the whole "just use a tight loop that do not yield to make the whole thing block")
<whitequark>
everything in the ocaml ecosystem currently uses lwt
<whitequark>
(except those parts which use async but I pretend they don't exist)
<flux>
well, you can still use LWT I think
<whitequark>
and this is nice, because you don't have five different incompatible ways to do concurrency
<whitequark>
as in, there aren't really any libraries written around blocking IO with Thread.
<Drup>
(I'm pretty sure you can implement lwt in terms of fibers and make the whole thing compatible)
<flux>
of course, then there is the chance that programs written to use LWT may not be really compatible with the non-co-operative task switching
<flux>
but instead they may assume that code without interruption points is executed atomically
<flux>
how is that going to be compatible with real threading?
<flux>
I do hope the folks introduce more concurrency primitives than just MVar.take and put, though, but I suppose that's given
<Drup>
flux: there's a question about that at the end
<flux>
(for example waiting for .taking and/or .putting multiple MVars would seem critical to me)
<Drup>
(by the parmap author, so not completely innocent :D)
<whitequark>
I fear the system with fibers will be more broken than the current state :?
<Drup>
why ?
<whitequark>
doesn't seem to map / interoperate well with existing usage
<Drup>
ah, well
<Drup>
raise your concerns to stephen dolan
<whitequark>
well
<Drup>
personally, I would just be happy to have parallelism and concurency properly integrated
<whitequark>
I assume that if those people are smart enough to make a multicore runtime, they're smart enough to not screw up integration
<whitequark>
so my concerns are probably unfounded
<Drup>
:p
<Drup>
I consider ocamllabs people quite smart in general =')
<libertas>
hi, I want to start witha functional PL, but have a question about something important to me.
fraggle_ has quit [Remote host closed the connection]
<libertas>
in C++, I can traverse a vector and change element with different functions according to the values of elements
<Drup>
libertas: don't ask to ask :)
<libertas>
and it happens in place, so the data is not copied
<Drup>
yes, you can do that in ocaml
<libertas>
but I read that ocaml needs to copy data
<libertas>
because its immutable
<Drup>
not everything is immutable
<Drup>
it's immutable *by default*
<ggole>
OCaml provides mutable records, strings and arrays
<Drup>
but you can use mutability when you feel it needed
<libertas>
and its speed is equivalent to C++?
<libertas>
I don't mind +10% of delay
tane has quit [Quit: Verlassend]
<Drup>
precise speed comparison is a game I don't like to play, but let's say it's in the same order of magnitude
mort___ has joined #ocaml
<libertas>
ok, that's enough for me
<Drup>
(and also, C++ speed depends a lot of the coding style)
<libertas>
of course
<whitequark>
the GC write barrier (invoked at every mutation) can be unexpectedly slow compared to C/C++
<whitequark>
but in general it's not that bad unless your workload is heavily dominated by mutation
philtom has joined #ocaml
<libertas>
in my case I'm building a clustering algorith for datastreams, so speed is important
fraggle_ has joined #ocaml
<Drup>
but mutability is not necessarily the good answer :)
<libertas>
when appling the same function to all elements of a vector, I guess we can use some 'map' function
<libertas>
with different functions I don't know if we can do it as well in OCaml
<whitequark>
you can. List.map or Array.map
<Leonidas>
BatList.map/Core_list.map %)
mort___ has quit [Quit: Leaving.]
shinnya has quit [Ping timeout: 264 seconds]
<libertas>
I'm also looking forward for concurrency and parallelism support which seem near
<Drup>
I don't know if it's idiomatic C, but it's very painful to make a typesafe binding that doesn't remove some interesting features
<adrien>
not really idiomatic C imho
<adrien>
well
<adrien>
there's nothing such as idiomatic C
<adrien>
but I found the API to be a bit annoying
zpe has joined #ocaml
<Drup>
what annoys me is the whole "you can pass MDB_FOO to some function to do something widely different, but you have the right to do that only if you passed MDB_BAR at the db creation, or the env opening, or somewhere else"
<whitequark>
oh :/
<whitequark>
no, it's not really very idiomatic C, it's just poor API
<Drup>
it's especially visible for the cursor operations
<Drup>
whitequark: do you think I should really expose cursor things or just some iter/map functions ?
<whitequark>
both
<whitequark>
cursor things as unsafe_* API
<Drup>
it's not really unsafe
<Drup>
just annoying
Thooms has joined #ocaml
<whitequark>
annoying_*
<Drup>
x)
zpe has quit [Ping timeout: 260 seconds]
<orbitz>
Drup: phantom types!
<Drup>
orbitz: well
<Drup>
I would, if it was only one or two things
<Drup>
but I don't want to have 6 or 7 phantom parameters >_>
<Drup>
especially for stuff that is only a configuration parameter
<orbitz>
ah i see
<orbitz>
it didn't realize it was a product
<orbitz>
off to explore the city, have a good night.
<Drup>
I do have a phantom type for read-only or write transaction, though
toroidalcode_ has joined #ocaml
toroidalcode has quit [Ping timeout: 252 seconds]
<Drup>
to prevent you from being stupid and write during a read only transaction
<Drup>
because that's quite common
pgomes has quit [Quit: Leaving]
<whitequark>
hahaha caml-list
<whitequark>
"I must say there's a strange feeling about creating more type definitions while breaking the type system."
sivteck has left #ocaml ["To Infinity and Beyond."]
cdidd_ has joined #ocaml
<Drup>
whitequark: @encoding doesn't seem to be documented
<whitequark>
Drup: there's a paragraph at the bottom
hhugo has joined #ocaml
teiresias has joined #ocaml
def` has joined #ocaml
asmanur has joined #ocaml
<Drup>
oh, indeed. It's easy to miss
<Drup>
not sure why you are using the poly variant syntax here, though
<whitequark>
I ... don't know actually
<Drup>
:D
zarul has joined #ocaml
zarul has joined #ocaml
zarul has quit [Changing host]
<whitequark>
I did put some thought into it back when I made it
<whitequark>
don't remember any of it whatsoever
hhugo has quit [Quit: Leaving.]
<Drup>
in case you don't want a normal identifier, for whatever reason, "String" seem better to me
fold has quit [Ping timeout: 255 seconds]
<Drup>
or `String, if you want to be consistent with YoJson
mfp has joined #ocaml
<Drup>
but `string seems akward
<whitequark>
nono, strings are off
<whitequark>
that's just dirty
<whitequark>
it should be some kind of variant
<whitequark>
String, `String or `string
<Drup>
yes, I meant String
<pippijn>
`string is deprecated, I thought
<Drup>
yes
<Drup>
hum
<Drup>
I think
<Drup>
whitequark: you can use "string" too, it will just be an identifier
Submarine has joined #ocaml
<whitequark>
Drup: it was the first thing I used
<whitequark>
but it's kinda horrible, it looks like a variable
<whitequark>
but it is not
<whitequark>
I hate such non-semantic use of AST, it should die in a fire
<Drup>
to me [@encoding string] look nothing more than an annotation
<whitequark>
other annotations include actual code though.
<whitequark>
[@encoding string] [@printer foobar]
<whitequark>
ewwwww
<Drup>
ok, that's a good argument
<Drup>
use String
<Drup>
:D
<whitequark>
well
<whitequark>
String is a module name.
<whitequark>
but `String is completely unambiguous
<Drup>
String is a variant
<Drup>
you can't put a module name there
<whitequark>
a longident :p yes, I know, but you also have a module named String, which is also a longident
<Drup>
I mean the good answer is to be consistent with YoJson
<Drup>
the json type contains `String, so use that
<Drup>
and in the future, you may add other encoding (for example, for list-like data structure that you want to be encoded like lists), and you could keep the same technique
<Drup>
s/mean/string/
<Drup>
graa
<Drup>
I should not multi task
<Drup>
I keep crossing the flows of information in my brain >_>
<whitequark>
hrm
<whitequark>
I was using the types of JSON itself there
samrat has quit [Quit: Computer has gone to sleep.]
msaegesser has quit [Read error: Connection reset by peer]
msaegesser has joined #ocaml
zpe has joined #ocaml
ygrek has quit [Ping timeout: 268 seconds]
zpe has quit [Remote host closed the connection]
zpe has joined #ocaml
msaegesser has quit [Ping timeout: 276 seconds]
zpe has quit [Ping timeout: 245 seconds]
axiles has quit [Remote host closed the connection]
Submarine has quit [Remote host closed the connection]
zpe has joined #ocaml
msaegesser has joined #ocaml
hhugo has joined #ocaml
robink_ is now known as robink
ggole has quit []
shinnya has joined #ocaml
Kakadu has quit [Quit: Konversation terminated!]
hhugo has quit [Quit: Leaving.]
zwer_ has joined #ocaml
zwer_ has left #ocaml [#ocaml]
darkf has joined #ocaml
zpe has quit [Remote host closed the connection]
zpe has joined #ocaml
zpe has quit [Ping timeout: 272 seconds]
msaegesser has quit [Ping timeout: 245 seconds]
MercurialAlchemi has quit [Ping timeout: 260 seconds]
msaegesser has joined #ocaml
Arsenik has joined #ocaml
rz has joined #ocaml
Arsenik has quit [Ping timeout: 246 seconds]
Simn has quit [Quit: Leaving]
zpe has joined #ocaml
zpe has quit [Ping timeout: 272 seconds]
rand000 has quit [Quit: leaving]
madroach has quit [Ping timeout: 250 seconds]
hhugo has joined #ocaml
madroach has joined #ocaml
<philtom>
how does one try the modular implicits branch? From watching the very blurry slides in the video I saw something like:opam switch 4.02.0+modular-implicits
<philtom>
But: [ERROR] "4.02.0+modular-implicits" is not a valid compiler.
NoNNaN has quit [Remote host closed the connection]