<appleby>
Ocaml is the worst programming language I have ever learned
<Drup>
appleby: if you are here around EU TZ daytime, adrien is the current lablgtk maintaner, he as a tutorial around, but I don't find the link already. He will be happy to answer questions.
<appleby>
k
<Drup>
appleby: if you are only used to OO programming, it's normal to find it difficult, don't blame the language =)
<appleby>
why not
<appleby>
the language is terrible lol
<r0bgleeson>
i like it
<appleby>
there's way too much random unnecessary syntax
<r0bgleeson>
i came from ruby
<r0bgleeson>
the syntax isn't pretty
<S11001001>
appleby: lol
<r0bgleeson>
some of the concepts are nice
<appleby>
and too many paradigms
<appleby>
like it tries to have features of every language or something and ends up like
<Drup>
appleby: for which language do you come from ?
<appleby>
like say I wanted to cook a dish of pasta
<S11001001>
no that's scala, appleby
<appleby>
ocaml is like
<appleby>
putting pasta+rice+tortilla chips+corn
<appleby>
and then spaghetti sauce and salsa and butter
<appleby>
and fish and ham and beef
<appleby>
all into one pot
<Drup>
don't try to find default in a language you now almost nothing about
<r0bgleeson>
your experience with a library shouldn't reflect on the language?
<Drup>
know*
<zRecursive>
ocaml is good at creating another language though :)
<r0bgleeson>
GTK+ is horrible in C as well from what i remember
<r0bgleeson>
probably slightly horrible anywhere you use it
<zRecursive>
because ocaml has mixed many features from other language
<Drup>
and you call ocaml a mess, you have interesting standards
<r0bgleeson>
appleby: i don't know your experience level but i learnt a lot by just reading through some of those chapters
<Drup>
appleby: so, let me be clear. If you want to ask question, go on, we will answer. If you try to troll this community by blaming the language for your own incompetence, you can find the door by yourself.
Puff has quit [Disconnected by services]
<appleby>
so
BitPuffi1 has joined #ocaml
<appleby>
the ocaml site tutorial
<appleby>
for introduction to gtk suggest you read about gtk 1.2 for C
<appleby>
I feel like my project uses gtk 2.6
<zRecursive>
The door is open if you are competence again :)
<appleby>
so is there a better way to go about getting a handle on lablgtk
mcclurmc has joined #ocaml
<Drup>
where do you read 1.2 ?
<appleby>
from the link from the introduction to gtk tutorial
<ygrek>
nice example of how _not_ to write (and build) bindings
mcclurmc has quit [Ping timeout: 252 seconds]
appleby has joined #ocaml
<appleby>
let ratio x y =
<appleby>
Float.of_int x /. Float.of_int y
<appleby>
;;
<appleby>
oh sorry
<appleby>
wrong place
S11001001 has quit [Quit: ERC Version 5.3 (IRC client for Emacs)]
siddhart1v is now known as siddharthv_away
<appleby>
# open Core.Std;;
<appleby>
Error: Unbound module Core
<appleby>
any ideas
appleby has quit [Quit: appleby]
siddharthv_away is now known as siddhart1v
zpe has joined #ocaml
zpe has quit [Ping timeout: 252 seconds]
mcclurmc has joined #ocaml
cdidd has quit [Ping timeout: 272 seconds]
mcclurmc has quit [Ping timeout: 252 seconds]
cdidd has joined #ocaml
wwilly has joined #ocaml
<flux>
I suppose there's no way to write this without the 'a parameter? type ('a) tinyg_session = ((<set : unit -> unit; ..>) as 'a) ?
avsm has joined #ocaml
zRecursive has joined #ocaml
axiles has joined #ocaml
axiles has quit [Client Quit]
avsm has quit [Ping timeout: 272 seconds]
wwilly has quit [Quit: This computer has gone to sleep]
Simn has joined #ocaml
ygrek has quit [Ping timeout: 248 seconds]
clan has quit [Quit: clan]
zpe has joined #ocaml
nikki93_ has joined #ocaml
BiDOrD_ has joined #ocaml
zpe has quit [Ping timeout: 252 seconds]
nicoo_ has joined #ocaml
ccasin_ has joined #ocaml
siddhart2v has joined #ocaml
jlouis has joined #ocaml
dant3 has joined #ocaml
mcclurmc has joined #ocaml
tlockney has quit [*.net *.split]
s4muel has quit [*.net *.split]
nikki93 has quit [*.net *.split]
mfp has quit [*.net *.split]
bernardofpc has quit [*.net *.split]
thorsten1 has quit [*.net *.split]
jlouis_ has quit [*.net *.split]
hcarty has quit [*.net *.split]
BiDOrD has quit [*.net *.split]
ccasin has quit [*.net *.split]
Armael_ has quit [*.net *.split]
vbmithr has quit [*.net *.split]
nicoo has quit [*.net *.split]
siddhart1v has quit [*.net *.split]
vbmithr has joined #ocaml
mcclurmc has quit [Ping timeout: 252 seconds]
s4muel has joined #ocaml
tlockney has joined #ocaml
hcarty has joined #ocaml
mfp has joined #ocaml
Armael_ has joined #ocaml
nickmeharry has joined #ocaml
thorsten1 has joined #ocaml
nickmeharry has quit [Client Quit]
bernardofpc has joined #ocaml
Kakadu has joined #ocaml
ygrek has joined #ocaml
demonimin has joined #ocaml
nicoo_ is now known as nicoo
jao has quit [Ping timeout: 252 seconds]
Asmadeus has quit [Quit: Changing server]
Asmadeus has joined #ocaml
yacks has quit [Quit: Leaving]
mcclurmc has joined #ocaml
Asmadeus has quit [Changing host]
Asmadeus has joined #ocaml
zRecursive has quit [Remote host closed the connection]
mcclurmc has quit [Ping timeout: 272 seconds]
<adrien>
bernardofpc: don't you get an output when doing #require? here it shows me the list of loaded libraries
yacks has joined #ocaml
IbnFirnas has joined #ocaml
nikki93_ has quit [Remote host closed the connection]
dant3 has quit [Read error: Connection reset by peer]
dant3 has joined #ocaml
zpe has joined #ocaml
zpe has quit [Ping timeout: 245 seconds]
mcclurmc has joined #ocaml
mcclurmc has quit [Ping timeout: 272 seconds]
jonludlam has joined #ocaml
vpm has quit [Ping timeout: 252 seconds]
<bernardofpc>
adrien: yes, I do
<bernardofpc>
it loads /usr/lib64/ocaml/lablgtk2/gtkInit.cmo:, /usr/lib64/ocaml/lablgtk2/gtkThread.cmo: /usr/lib64/ocaml/lablgtk2/lablgtk.cma: (in the opposite order)
arj has joined #ocaml
Thooms has joined #ocaml
jonludlam has quit [Ping timeout: 246 seconds]
jonludlam has joined #ocaml
<adrien>
hmm
<adrien>
so what changed after 1:50am?
<bernardofpc>
I wrote explicitly #load "gtkThInit.cmo" ;; on the toplevel
mort___ has joined #ocaml
<bernardofpc>
and this made the UI elements appear immediatly after creating them, and moreover gave me toplevel prompt
<bernardofpc>
brb
<adrien>
hmm
<adrien>
shouldn't happen
<adrien>
well
<adrien>
#threads;; after #use "topfind";;
<adrien>
should be enough
zpe has joined #ocaml
zpe has quit [Read error: Connection reset by peer]
lostcuaz has joined #ocaml
ollehar has joined #ocaml
dant3 has quit [Remote host closed the connection]
<arj>
wasn't there an opam website where all the package's docs/apis where published?
<arj>
Kakadu: I was looking for a page showing the ocamldocs, but maybe I am mistaken, because I just found an opam issue stating that this is not easy.
<Kakadu>
Ah, sorry
<Kakadu>
I ahve read your message incorrectly
ggole has joined #ocaml
thomasga has joined #ocaml
troydm has quit [Quit: What is hope? That all of your wishes and all of your dreams come true? (C) Rau Le Creuset]
NoNNaN has quit [Ping timeout: 240 seconds]
NoNNaN has joined #ocaml
lostcuaz has quit [Read error: No route to host]
lostcuaz has joined #ocaml
mcclurmc has joined #ocaml
csakatoku has quit [Remote host closed the connection]
<bernardofpc>
adrien: I did use (topfind, threads, gtk-init) and it did not load... maybe because everything was installed without opam (I have no opam in my distro :/)
Simn has quit [Ping timeout: 252 seconds]
csakatoku has joined #ocaml
thomasga has quit [Quit: Leaving.]
thomasga has joined #ocaml
Simn has joined #ocaml
nikki93 has quit [Ping timeout: 250 seconds]
lostcuaz has joined #ocaml
vpm has joined #ocaml
siddhart2v has quit [Quit: leaving]
FreeArtMan has joined #ocaml
siddharthv has joined #ocaml
kaustuv has joined #ocaml
ollehar has quit [Quit: ollehar]
_andre has joined #ocaml
AltGr has joined #ocaml
<arj>
bernardofpc: you could install opam yourself?
<bernardofpc>
I could, but I'm not really keen on doing that
<bernardofpc>
and if some package has a different behaviour according to being installed with/without opam, that should be documented
appleby has joined #ocaml
<arj>
oh, that's really bad. I wasn't aware of the context.
lostcuaz has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<bernardofpc>
arj: I'm not saying that's the case !
<bernardofpc>
I'm raising that as a hypothesis because everyone indicated only 3 commands, and turned out a 4th was needed
ohama has joined #ocaml
yacks has joined #ocaml
<arj>
ah, ok, well then I fully messed up, sorry. Only read half of the discussion, shouldn't do that.
<bernardofpc>
no pb ;-)
<adrien_oww>
bernardofpc: I don't use opam and it works for me
Yoric has quit [Remote host closed the connection]
<adrien_oww>
Kakadu: but it's nicer to avoid -rectypes
<adrien_oww>
I'll let someone explain how :P
<flux>
IIRC the reason is that benign looking code can produce the most interesting error messages, and some code that the developer might expect not to compile will compile..
<flux>
also it's not the standard way ;-)
<companion_cube>
smondet: I'm not sure what SML does
Yoric has joined #ocaml
<flux>
I suppose no word about thread support in merlin.. I really should make a github issue of it 8-)
<adrien_oww>
flux: and everything which depends on something that uses -rectypes needs to be built with -rectypes too
<flux>
actually I also have trouble with Merlin not always finding the proper place to highlight, I think it's because of tabs in the code..
<flux>
other than those two things, merlin is super :)
cthuluh_ is now known as cthuluh
patronus has joined #ocaml
mrvn_ is now known as mrvn
<adrien_oww>
well, how can you count chars properly with variable-width indent? :P
<flux>
yeah, we don't yet have technology for that..
AltGr has joined #ocaml
<companion_cube>
we have, it's called "space-based indentation" ;)
mcclurmc has joined #ocaml
<bernardofpc>
companion_cube: like in column-determined program meaning ?
<bernardofpc>
(hint: it has 66 but it's not a route)
mcclurmc has quit [Ping timeout: 245 seconds]
axiles has joined #ocaml
nikki93 has quit [Ping timeout: 245 seconds]
sgnb` is now known as sgnb
ygrek has joined #ocaml
lostcuaz has joined #ocaml
sgnb is now known as Guest13094
Guest13094 is now known as sgnb`
dant3 has quit [Remote host closed the connection]
<companion_cube>
noes, not fortran :D
<companion_cube>
I just meant indenting with spaces, of course
sgnb` is now known as sgnb
dant3 has joined #ocaml
<gasche>
adrien_oww: -rectypes only propagates through interfaces
<gasche>
(I think)
<gasche>
so it's safe to use -rectypes only in an implementation
<gasche>
flux: the editor-side frontend of merlin could probably re-process the relevant line and adjust column number accordingly
<gasche>
do open an issue on this one
<rks>
or better: don't mix tab and spaces
<rks>
:)
<gasche>
if possible, the tool should adapt to the users and not otherwise
<rks>
:-°
<companion_cube>
gasche: I'm not sure why you use continuations everwhere in your interface
<gasche>
companion_cube: to preserve tail-recursivity
<companion_cube>
I mean, why Acc.run instead of Acc.create, for instance
<gasche>
it's nicer because I don't have to provide a ('a mut_list -> 'a list) function
<companion_cube>
otoh the latter is simpler
<companion_cube>
for instance you can have a 'a mut_list list and convert it into a 'a list list
<companion_cube>
(it should even be a primitive since it's actually Obj.magic)
<gasche>
... but that's what we already have in today's Accum
<gasche>
the point is to show that it's possible to completely hide dummy elements from the interface
<gasche>
(or any creating function really)
<adrien_oww>
gasche: I'd think so but I have some doubts too
<companion_cube>
gasche: why would you have to show dummy elements?
ontologiae_ has quit [Ping timeout: 245 seconds]
yacks has quit [Ping timeout: 272 seconds]
dch has joined #ocaml
<dch>
I’m looking for any high-traffic UDP-based library to read through & understand — anything to recommend?
<companion_cube>
the only UDP things I can think of, in Ocaml, would be 1/ a DNS implementation (forgot the name) 2/ mldonkey (p2p application)
mort___ has joined #ocaml
martintrojer has joined #ocaml
<Drup>
for the 1), if we are thinking the same, it's in the mirage project
ocp has joined #ocaml
<adrien_oww>
0/ use lwt
<adrien_oww>
(or async if you prefer it)
<companion_cube>
there is a "dns" package on opam
martintrojer has quit [Quit: ERC Version 5.3 (IRC client for Emacs)]
<companion_cube>
adrien_oww: I think the point here is reading code, not writing
<adrien_oww>
read lwt's examples :P
mcclurmc has joined #ocaml
nikki93 has joined #ocaml
mcclurmc has quit [Ping timeout: 246 seconds]
yacks has joined #ocaml
lostcuaz has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
darkf has quit [Quit: Leaving]
<bernardofpc>
gasche: partition long 15.14s -> "after specializing" runs slower than the previous one !
<rixed>
J'arrive a retourner la bas (en fr) tous les 2 WE. Ca coute pas trop cher et c'est assez pratique (loue une voiture à beauvais)
<rixed>
/o\ wrong chan :)
dant3 has quit [Remote host closed the connection]
sagotch_ has joined #ocaml
<sagotch_>
Hi everyone.
<sagotch_>
Someone could tell me why adding "-lib str" to my ocamlbuild line gives me "ocamlfind ocamlopt -a str.cmxa src/typeTo.cmx src/parsetoml.cmx src/tomlex.cmx src/to.cmx -o src/to.ml.cmxa Option -a cannot be used with .cmxa input files." ?
<bernardofpc>
out of curiosity, if I have a record r whose fields are also records, and I do let r' = {r with x = blah} does this copy (as in "allocates memory for") all other registers from r into r' or not ?
AltGr has left #ocaml []
AltGr has joined #ocaml
<ggole>
{ r with ... } allocates one record.
<ggole>
There's no deep copying in the current implementation.
<bernardofpc>
ok, thanks !
<flux>
hmm
<flux>
so it does copy all the fields of the original record, as it needs to allocate a new one. but the copying is only done shallowly.
dant3 has joined #ocaml
<flux>
so for example if both have a string field, after a 'copy' they will share the string field; however, if the string field itself is mutable, the field itself can be changed to point to a new string without affecting the other
<ggole>
That's not necessarily positive, of course.
<ggole>
(For performance: the semantics are both reasonable and clear.)
<gasche>
bernardofpc: that's probably noise
<gasche>
sagotch_: -use-ocamlfind -package str
zpe has quit [Remote host closed the connection]
<gasche>
your example seems to indicate that -lib may be slightly buggy with respect to .cmxa creation
<companion_cube>
hmmm, is there a way to define a type 'a t = ..... where 'a must be a subtype of foo?
<gasche>
companion_cube: constraint?
<companion_cube>
what would it look like?
zpe has joined #ocaml
<sagotch_>
@gasche awesome! thank you
<companion_cube>
I only found equality constraints
<bernardofpc>
gasche: probably, but it'scurious that it's so big
<gasche>
companion_cube: class type foo = object .. end type 'a t = ... constraint 'a = #foo
<companion_cube>
arf, but I don't use objects
<companion_cube>
(it's subtyping via private aliases)
<sagotch_>
(and why lib works for cma file, but not for cmxa ?)
<gasche>
sagotch_: .cma can contain other .cma, .cmxa cannot contain .cmxa
dant3 has quit [Remote host closed the connection]
dant3 has joined #ocaml
<gasche>
(generally bytecode objects are in an OCaml-friendly format that can be easily manipulated, while native objects are ELF/whatever stuff and it's non-portable to generate them, and even less reasonable to inspect them after the fact)
zpe has quit [Ping timeout: 246 seconds]
talzeus has joined #ocaml
<sagotch_>
so, my question should be: what is the difference link with -lib or -package ? (can't find doc about these, and help says link with ocaml library / findlib package)
<gasche>
in the old times of history, ocamlbuild had no special support for ocamlfind (because old-timers among which it was developped didn't use it)
thomasga has quit [Quit: Leaving.]
<gasche>
so ocamlbuild embeds a fair bit of domain-specific logic about archives, how to link them and when, etc., embodied by the -lib option (and it's still useful when the project locally builds library archives and then compiles other stuff on top of it)
<adrien_oww>
because ocamlbuild was created before most projects used ocamlbuid
<gasche>
but nowadays, for any *external* dependencies, it's much simpler to just rely on ocamlfind to do The Right Thing
dant3 has quit [Ping timeout: 272 seconds]
<gasche>
which is what -package does : it passes the option "-package foo" to ocamlfind and voilà, job done
<gasche>
it therefore subsumes -lib
<sagotch_>
You means that both command do the same thing, but not the same way?
<sagotch_>
okay
<gasche>
some people still don't want to depend on ocamlfind so -lib is still useful for them
<sagotch_>
thanks for explaining
<adrien_oww>
s/some people/compiler people/
<adrien_oww>
which makes sense
<adrien_oww>
others, we can convince them
<companion_cube>
but they don't really need libraries, do they?
<companion_cube>
the compiler people
<adrien_oww>
but you're still using "ocamlfind ocamlc foo.ml" instead of "ocamlc foo.ml"
<companion_cube>
indeed
<adrien_oww>
but maybe a warning if you're using "-lib" and no "-without-ocamlfind" (or whatever it is called)
<bernardofpc>
--I-know-what-I-am-doing :D
<bernardofpc>
(although I like -lib because it's shorter when it works)
<bernardofpc>
(of course, if --use-ocamlfind becomes default, this point goes away, -packages=... or -libs=... is not so much different anymore)
thomasga has joined #ocaml
bfn__ has joined #ocaml
<adrien_oww>
bernardofpc: it is now the default
mcclurmc has joined #ocaml
bfn___ has joined #ocaml
bfn__ has quit [Ping timeout: 245 seconds]
FreeArtMan has quit [Remote host closed the connection]
dant3 has joined #ocaml
FreeArtMan has joined #ocaml
shinnya has quit [Ping timeout: 272 seconds]
appleby has joined #ocaml
<gasche>
adrien_oww: it's not
<gasche>
this was discussed but it was considered too invasive a change before the 4.01 deadline
<gasche>
(and it's not only the compiler people, as you remember Mirage people are worried about ocamlfind's build time0
<adrien_oww>
gasche: isn't it in for 4.02?
<adrien_oww>
which is not yet released, yeah
<adrien_oww>
gasche: I'd still put the warning :P
<gasche>
I haven't done much ocamlbuild work since the release, busy with other things
<adrien_oww>
but there's findlib and I'd probably prefer to see a findlib-provided plugin to ocamlbuild
<adrien_oww>
that requires other plumbing
<adrien_oww>
though
siddharthv is now known as siddharthv_away
<gasche>
appleby: re. your core question, you should have a look at the "installing the things you need" chapter of Real World OCaml, if that's what you're using as a learning document
<gasche>
(otherwise: don't use Core if you're a beginner)
<gasche>
(and not using RWO, I mean)
yacks has quit [Ping timeout: 250 seconds]
nikki93 has joined #ocaml
appleby has quit [Quit: appleby]
ygrek has quit [Remote host closed the connection]
ygrek has joined #ocaml
sagotch_ has quit [Quit: Page closed]
Dongyancai has joined #ocaml
appleby has joined #ocaml
<appleby>
gasche
<appleby>
i'm using real world ocaml of course
<appleby>
and I did opam install core utop already
<kaustuv>
Does it do anything that the toplevel + emacs does not provide?
Yoric has quit [Ping timeout: 245 seconds]
<Drup>
type aware auto completion
<Drup>
hum
<Drup>
not type aware, actually, context aware
<kaustuv>
What does that mean? In emacs I have M-x hippie-expand that completes based on "nearby" text (which could be text in other windows or frames or buffers as well)
<Drup>
that's just statistical
appleby has quit [Quit: appleby]
<ggole>
I like emacs' "dumb" completion: it often picks up things in comments that "smart" completion doesn't.
<ggole>
Or in strings, or in temp buffers, etc
<Drup>
I don't like when "completion" pick stuff in buffers and other text source
<Drup>
because it's always stupid and that's not what I want
mort___ has joined #ocaml
Kakadu has quit [Ping timeout: 245 seconds]
<ggole>
OK, but "utop is great because it does what I personally like" isn't much of an argument if I my tastes happen to differ
nikki93 has quit [Ping timeout: 250 seconds]
<sagotch>
Does someone have a good link about installing its own library ?
zpe has joined #ocaml
<sagotch>
It tried ocamlfind install and see my library use topfind list in toplevel, but cannot use it nor compile a file opening it.
<companion_cube>
did you install all the files? META, the .cmi, .cma, .cmxa, .a...
<sagotch>
only META cmxa and cma
<Drup>
ggole: well, you test it and you like it or not :D
<rks>
utop also provides specific support for async (and lwt?) iirc
<rks>
you have a #typeof directive
<ggole>
Drup: yup, fair enough
<flux>
utop is very.. flashy. but at least one doesn't need to use rlwrap (if not using emacs)
<sagotch>
`ocamlfind list` gives me the line "toml (version: 20140204)", but using open Toml in a .ml produce unn=bound module error
<flux>
rks, nice, thanks :)
<rks>
my pleasure
<sagotch>
and I try to compile using `ocamlbuild -use-ocamlfind -package toml toto.byte`
<Dongyancai>
Hello, hope here is someone using ocamlnet. I have installed the ocamlnet-3.6.3 but don't know how to make the fcgi example usable. it says Fatal error: exception Unix.unix_error(40, "accept", "")
<flux>
something is probably trying to use accept on a Unix.file_descr that's not a socket
thomasga has joined #ocaml
<ggole>
Printing of constructors is such a sketchy part of ocaml :/
jao has joined #ocaml
<ggole>
I guess it's not really the language proper that is to blame. Still, yuck.
jao has quit [Changing host]
jao has joined #ocaml
<flux>
dongyancai, btw, be sure you compile with -g and then export OCAMLRUNPARAM=b
<flux>
if it's unclear where the accept is being called from
<flux>
but probably it's not..
dant3 has quit [Remote host closed the connection]
<Dongyancai>
I'm compiling with "ocamlfind ocamlc -o add.fcgi -package netcgi2 -linkpkg add.ml add_fcgi.ml" as the example says.
<flux>
are you executing it manually?
<flux>
I think it's supposed to be executed by the web server
<flux>
fcgi protocol is so weird. I like prefer if I can use it.
<Dongyancai>
Yes, manually, as I tried C version hello world of fastcgi and it ran manually...
lostcuaz has joined #ocaml
<flux>
and did what? I don't think it's possible to interact with fastcgi from console..
Yoric has joined #ocaml
<flux>
I think the C version just had a fallback to non-fcgi, whereas ocamlnet doesn't (unless you make it yourself)
<Dongyancai>
No interaction, i think it should work as a deamon, but seems to work well if start manually.
<flux>
and did it work from a web server?
<flux>
because I think it probably did not use that instance for the work.. "The Web server leaves a single file descriptor, FCGI_LISTENSOCK_FILENO, open when the application begins execution. This descriptor refers to a listening socket created by the Web server." so it inherits a socket it listens to. (though I do wonder what ocamlnet is doing if the environment variables is not set! accepts fileno 0?)
<flux>
scgi works like you describe: you can start it separately, and then the server connects it to do the work
<flux>
as I think it has fallen back to CGI mode as the page suggests: "If the application was invoked as a CGI program, the first call to Accept returns 0 and the second always returns -1, producing CGI behavior. (See "FCGI_Accept (3)" on page 21 for details.) "
<Dongyancai>
thank you, I'll watch on it.
<flux>
good luck!
ygrek has quit [Ping timeout: 250 seconds]
Anarchos has joined #ocaml
Kakadu has joined #ocaml
S11001001 has joined #ocaml
S11001001 has quit [Changing host]
S11001001 has joined #ocaml
lucu_duort has joined #ocaml
<kaustuv>
Regarding completion, I've been using merlin with auto-complete for a few weeks now, and I have to say that on the whole auto-complete is a big annoyance. I find the constant context-switching from "writing" to "choosing" modes disorienting. Does it actually increase anyone's produtivity?
<rks>
kaustuv: you could try « completion-at-point »
<kaustuv>
That doesn't seem to be much of an improvement. I actually like C-c TAB a lot. I just wish auto-complete would shut up about completions all the time and wait for me to hit C-c TAB.
<kaustuv>
Yes, I should probably have tried to look up how to configure it instead of bitching on IRC, but it's so much easier to bitch
<companion_cube>
^^
<def-lkb>
and yes, by default auto-complete is a complete annoyance
<gasche>
maybe disabling it would be the right default
ygrek has joined #ocaml
<Dongyancai>
flux, thank you very much. I thought it does the listening itself, but indeed, ocamlnet also needs spawn-fcgi.
<kaustuv>
def-lkb: thanks, niling ac-start-auto did the trick
rz has joined #ocaml
nikki93 has joined #ocaml
zpe has quit [Ping timeout: 272 seconds]
Yoric has quit [Ping timeout: 245 seconds]
mort___ has quit [Ping timeout: 272 seconds]
talzeus has quit [Read error: Connection reset by peer]
lucu_duort has quit [Ping timeout: 245 seconds]
sagotch has quit [Ping timeout: 245 seconds]
<kaustuv>
It looks like PR#5157 is the oldest report from me with no activity on it. The patch for it is literally trivial (https://gist.github.com/chaudhuri/8807864) and it seems silly to open a PR for it.
jwatzman|work has joined #ocaml
vkz has joined #ocaml
sagotch has joined #ocaml
thomasga has quit [Quit: Leaving.]
<vkz>
hi all. A really silly question. Where do I find docs for ocaml packages? ocaml.org says specific packages should link to their docs, but it doesn't appear so. For example Core: https://opam.ocaml.org/pkg/core/core.109.60.00/ has nothing
Drup has quit [Excess Flood]
fantasticsid has quit [Write error: Broken pipe]
Drup has joined #ocaml
dant3 has joined #ocaml
<kaustuv>
gasche: if you want to do patch reviews on github, you should maybe add a .gitignore to ocaml. Here's a brute-force version that you can probably pare down: https://github.com/chaudhuri/ocaml/commit/16d6e2d0
<gasche>
good idea, thanks
<gasche>
I'm currently looking at how your M.[foo] idea can be implemented in Camlp4
<gasche>
... and it's not pretty
<kaustuv>
My earlier patch had an answer for the pre-splitup camlp4...
nikki93 has quit [Ping timeout: 245 seconds]
<gasche>
oh
<gasche>
but did it work?
<gasche>
it looks wrong from a distance
<gasche>
..._dot_lparen assumes a parenthesis, and you close with other delimiters
<kaustuv>
ah, you're correct.
<gasche>
would you submit your .gitignore as a pull request?
<kaustuv>
done
<sagotch>
Hey question about strange behavior moving a declaration outside mutually recursive declaration blocks: http://pastebin.com/VcC5cm6R
<sagotch>
why does the version where string_of_list is defined inside the muttually recursive block gives a type error ?
<sagotch>
(hum... sorry, I have issues with my connection, so I may be unable to read answers....)
dant3 has quit [Remote host closed the connection]
dant3 has joined #ocaml
kaustuv has left #ocaml []
sagotch has quit [Ping timeout: 250 seconds]
<gasche>
sagotch: polymorphic recursion, you need an explicit annotation: and string_of_list : 'a . ('a -> string) -> 'a list -> string = fun stringifier els -> ...
<gasche>
oh well.
<gasche>
going home anyway
mreca has joined #ocaml
oriba has joined #ocaml
sagotch has joined #ocaml
<gasche>
sagotch: polymorphic recursion, you need an explicit annotation: and string_of_list : 'a . ('a -> string) -> 'a list -> string = fun stringifier els -> ...
<gasche>
(declaring it upfront makes let-generalization do the work for you)
<sagotch>
yep thanks, I was reading logs :)
<ggole>
Why the restriction? Soundness?
<sagotch>
I note it and will try to find some explanation on it
sagotch has quit [Remote host closed the connection]
yacks has quit [Ping timeout: 246 seconds]
skchrko has quit [Quit: Leaving]
companion_cube has quit [Read error: Operation timed out]
hto has quit [Read error: Operation timed out]
rixed has quit [Read error: Operation timed out]
rom1504 has quit [Read error: Operation timed out]
mehdid has quit [Read error: Operation timed out]
fayden has quit [Read error: Operation timed out]
jdoles has quit [Read error: Operation timed out]
johnelse has quit [Read error: Operation timed out]
jpdeplaix has quit [Read error: Operation timed out]
dkg has quit [Read error: Operation timed out]
mehdid_ has joined #ocaml
rom1504 has joined #ocaml
mehdid_ is now known as mehdid
jpdeplaix` has joined #ocaml
fayden has joined #ocaml
dkg has joined #ocaml
Anarchos has quit [Ping timeout: 250 seconds]
companion_cube has joined #ocaml
ygrek has quit [Ping timeout: 247 seconds]
rixed has joined #ocaml
hto has joined #ocaml
johnelse has joined #ocaml
jdoles has joined #ocaml
vkz` has joined #ocaml
zpe has joined #ocaml
vkz has quit [Ping timeout: 272 seconds]
zpe has quit [Ping timeout: 246 seconds]
tlockney has quit [Excess Flood]
tlockney has joined #ocaml
AltGr has left #ocaml []
AltGr has joined #ocaml
ollehar has joined #ocaml
Anarchos has joined #ocaml
AltGr has left #ocaml []
nikki93 has joined #ocaml
sagotch has joined #ocaml
appleby has joined #ocaml
nikki93 has quit [Ping timeout: 250 seconds]
thomasga has joined #ocaml
thomasga has quit [Client Quit]
sagotch has quit [Ping timeout: 250 seconds]
appleby has quit [Quit: appleby]
rz has quit [Quit: Ex-Chat]
mreca has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
wwilly has joined #ocaml
sagotch has joined #ocaml
skchrko has joined #ocaml
ggherdov_ has quit [Ping timeout: 264 seconds]
dsheets has quit [Ping timeout: 250 seconds]
vkz has joined #ocaml
nikki93 has joined #ocaml
ggherdov_ has joined #ocaml
Arsenik has joined #ocaml
ulfdoz has joined #ocaml
_andre has quit [Quit: leaving]
NaCl__ is now known as NaCl
NaCl is now known as Guest79996
Guest79996 is now known as NaCl__
Thooms has joined #ocaml
kulhas has joined #ocaml
<kulhas>
hi, I am trying ocaml, how can I match end of file with input_line. I saw that I can try catch the end of file exception but I would prefer to catch in match with. tx
<Kakadu>
exceptions are OK. They are not so heavy as JVM's ones
dsheets has joined #ocaml
<adrien>
kulhas: match and the try/with construct are separate; you cannot match an exception with match/with, only with try/with
<adrien>
and as Kakadu mentioned, there's no reason to try to avoid the
<adrien>
m
<kulhas>
ok tx, I will use the exception
Yoric has joined #ocaml
ggole has quit []
clan has joined #ocaml
<adrien>
gasche: bleh^2: List.sort_uniq is a recent addition and that's not mentioned in the documentation ='(
Yoric has quit [Remote host closed the connection]
<adrien>
(eventually I'll grep my IRC logs and make a patch)
arj has quit [Quit: Leaving.]
ocp has joined #ocaml
clan has quit [Quit: clan]
kulhas has quit [Quit: Leaving.]
mort___ has joined #ocaml
zpe has joined #ocaml
zpe has quit [Remote host closed the connection]
rz has joined #ocaml
zpe has joined #ocaml
zpe has quit [Read error: Connection reset by peer]
zpe has joined #ocaml
thomasga has joined #ocaml
Thooms has quit [Remote host closed the connection]
Thooms has joined #ocaml
vkz`` has joined #ocaml
zpe has quit [Remote host closed the connection]
zpe has joined #ocaml
vkz` has quit [Remote host closed the connection]
nikki93 has quit [Remote host closed the connection]
mort___ has quit [Quit: Leaving.]
appleby has joined #ocaml
zpe has quit [Ping timeout: 272 seconds]
nikki93 has joined #ocaml
clan has joined #ocaml
nikki93 has quit [Remote host closed the connection]
appleby has quit [Ping timeout: 256 seconds]
appleby has joined #ocaml
clan has quit [Client Quit]
ulfdoz has quit [Ping timeout: 250 seconds]
ulfdoz has joined #ocaml
nikki93 has joined #ocaml
<flux>
I have gcode.mllib with GCodeParser and GCodeLexer, gocdePaser.ml and gcodeLexer.mll, and ocamlbuild gcode.cma doesn't even try to compile. am I doing something obviously wrong?
<flux>
it tries to build gcode.cmo for some reason
<flux>
though I still don't understand why it tries to compile gcode.ml, which didn't exist
<flux>
leading to suberbly bad error message
<companion_cube>
must be the default behavior
shinnya has joined #ocaml
<flux>
something like that. I wouldn't mind if it once told me that it cannot find a file, instead of just 'Failed to build all of these'
<sagotch>
some questions about library writing: what do we **must** put in .mlllib file ?
<companion_cube>
the list of module names you want to put in the library
<sagotch>
module i want to put or module needed by others too ?
<companion_cube>
do you mean dependencies?
<sagotch>
I mean other module a I wrote used by the one I want to use
<sagotch>
I want to put in library*
<sagotch>
are they automatically packaged with my main file?
jsc has joined #ocaml
jsc has left #ocaml []
<companion_cube>
sagotch: you only need a .mllib if you want to create a library
mort___ has joined #ocaml
<sagotch>
hum... what actually is a library (what is the different with a .cm(x)a file)?
<companion_cube>
the same thing
<companion_cube>
I meant, by opposition to compiling a program
<flux>
.cm(x)a contains multiple .cmo files
<flux>
usually when you make something used for compiling other programs you make a library..
<flux>
(alternative: you can make a pack)
<flux>
(pack = multiple .cmo-file packed into one .cmo-file)
<sagotch>
ok
<sagotch>
so a library is actually what I need, but I found very few (not to say none) documentation about it. removing the .mllib file does not seems to change anything (is it a behavior I should expect ?)
nikki93 has quit [Remote host closed the connection]
<sagotch>
and bonus question: what to put inside a META file
<flux>
.mllib is a file used by ocamlbuild to know that you want to put many .cmo files into libraryname.cma
<flux>
META file, well it's actually well documented in findlib documentation
<flux>
but you might want to get an overview by just looking at into existing META files
<flux>
basically you put in name, dependencies, and the names of the library files
<flux>
and you want to ocamlfind install libraryname *.cma *.cmxa *.a *.cmi *.mli
<companion_cube>
sounds like you want to use oasis :)
<flux>
but, sleep
nikki93 has joined #ocaml
Yoric has quit [Remote host closed the connection]
Yoric1 has joined #ocaml
<sagotch>
I think I used META file properly, but I only can access function defined into the "main" module
<sagotch>
TypeTo is "known", since it is used in type definition
<sagotch>
but I did not found how to access this TypeTo module
<sagotch>
can't use constructor, can't use function.
<sagotch>
(TypeTo is in mllib file)
<sagotch>
and in META : [...] archive(byte) ="toml.cma"
dant3 has quit [Remote host closed the connection]
Kakadu has quit [Quit: Konversation terminated!]
nikki93 has quit [Remote host closed the connection]
Arsenik has quit [Remote host closed the connection]
vkz`` has quit [Ping timeout: 246 seconds]
vkz has quit [Ping timeout: 272 seconds]
<hnrgrgr>
sagotch: did you install typoTo.cmi together with toml.cma ?
<sagotch>
nope, only toml.* files
<hnrgrgr>
you have to. It contains the module signatures.
<hnrgrgr>
the .cma file only contains the bytecode.
<sagotch>
but why does toplevel knows type contained in TypeCode module?
<adrien>
is the .cmi file in the current directory?
<sagotch>
it is not
<hnrgrgr>
Quite unintuitive. It does rellay knows types contained in TypeCode. Try 'type alias = TypeTo.tomlTable'. Without typeTo.cmi, it will probably fail.
<hnrgrgr>
It does rellay -> it doesn't really
FreeArtMan has quit [Quit: main.lv]
<sagotch>
indeed, it fails
<hnrgrgr>
It is just saying something like 'the type of Toml.parse reference a type tomlTable in a module TypeTo'
<hnrgrgr>
or, better said, it just print the name of the type, without knowing the type.
<sagotch>
Did not know it could happen, thanks.
<sagotch>
I am providing cmi files in `ocamlfind install` line. I should do it somewhere else ?
<hnrgrgr>
should be ok.
<sagotch>
or is it the right way to di
<sagotch>
do*
<sagotch>
Everything seems right with cmi files provided, thank you.
zpe has joined #ocaml
<hnrgrgr>
great.
jonludlam has quit [Remote host closed the connection]
zpe has quit [Ping timeout: 246 seconds]
smondet has quit [Quit: leaving]
smondet has joined #ocaml
nikki93 has joined #ocaml
wwilly has quit [Quit: This computer has gone to sleep]
axiles has quit [Remote host closed the connection]
Thooms has quit [Quit: WeeChat 0.3.8]
ccasin_ is now known as ccasin
appleby has quit [Quit: appleby]
lostcuaz has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
zpe has joined #ocaml
sagotch has quit [Remote host closed the connection]
clan has joined #ocaml
nikki93 has quit [Remote host closed the connection]
nikki93 has joined #ocaml
clan has quit [Client Quit]
zpe has quit [Remote host closed the connection]
zpe has joined #ocaml
zpe has quit [Read error: Connection reset by peer]