<dmbaturin>
kakadu: opam-repository-mingw works fine for me. It's nothing like the pains of ML on windows from a few years ago.
<dmbaturin>
Of course I'd rather use opam-cross-windows and test in wine, but the cross-repository needs work.
mononote has left #ocaml ["Disconnected: closed"]
mononote has joined #ocaml
mononote has left #ocaml ["Disconnected: closed"]
mononote has joined #ocaml
Haudegen has quit [Quit: Bin weg.]
spew has joined #ocaml
<Leonidas>
can someone make ivg write a book? I would totally read it.
<Armael>
just ask questions on discuss and concatenate his answers
jao has joined #ocaml
tane has joined #ocaml
<dmbaturin>
Leonidas: My book needs contributors. ;)
<Leonidas>
Armael: this could absolutely work :)
<dmbaturin>
My effort to make a living document that doesn't age badly like RWOv1 did is happily ignored by everyone. I _will_ complete it eventually, but well...
webshinra has quit [Remote host closed the connection]
<tane>
dmbaturin, is there a draft available? :)
<hannes>
yes for a book of ivg's answers, I'd read it
Haudegen has joined #ocaml
<narimiran>
oh tane, you're here too :)
<tane>
narimiran, yeah, you asked here about that stuff, that's how I remembered :)
<narimiran>
:)
<narimiran>
yep, that's me
<narimiran>
but now i'm one month more experienced with ocaml :)
<tane>
nice
<Leonidas>
dmbaturin: RWOv1 is still relevant to some degree. What hasn't is the camlp4 bits.
<tane>
are you using vim/emacs?
<narimiran>
back to your comment: i usually don't do `open Containers`, i use CCArray.xxx CCList.xxx etc, so i can immediately see what is coming from where
<Leonidas>
dmbaturin: you can of course leave out ppx in your document to err on the safe side
<narimiran>
don't know if that's the "correct" way to do it, but that's what i've been doing
<Leonidas>
kinda like companion_cube and dbunzli avoid ppx altogether
<narimiran>
i use vim + merlin, and it is a great combo!
<tane>
narimiran, ah ok, well I like the "overlay", because I'd always either use containers or base
<tane>
but haven't written anything proper for a while, except for some prototypical implementations
<narimiran>
hmm, i've never checked if there's a way (in dune, i guess) to have automatic `open Foo` for every file in a project
<narimiran>
now, should i do `-open Containers` and just use all of its stuff with Array.xxx, List.xxx, etc., or is my current way of "CCArray.xxx, CCList.xxx, etc. only when necessary, otherwise use stdlib" better? does it matter?
<narimiran>
dmbaturin: i just skimmed through some chapter and two things immediately grabbed my attention:
<narimiran>
+ finally a website where i don't have to zoom in, it uses a large-enough font size
<narimiran>
- your lines are way too long. now the body is 1000px wide, try putting it around 850px to improve readability
dhil has joined #ocaml
<narimiran>
as for the content, i haven't had the time to read it yet, but i will do it
<narimiran>
- once you're reading some chapter, there's no link back to the index
<narimiran>
...nor to the next chapter
<dmbaturin>
There are sure details to take care of. Maybe the list of chapter should be in a sidebar really.
<dmbaturin>
* chapters
<dmbaturin>
Strong copyleft license is kinda the whole point. It belongs to the community, every can pick it up even if I completely stop caring.
<rgrinberg>
narimiran i'm not a fan of `-open` myself
<tane>
rgrinberg, the flag or `open X` in general?
<rgrinberg>
both
<tane>
for containers or base I open, otherwise I open locally
<tane>
or use the Mod.(..) syntax, which is nice too :)
dborisog has quit [Ping timeout: 240 seconds]
jco has joined #ocaml
<jco>
Hello!
<jco>
I have a module P with a type t, which is an integer
<jco>
then how do I pass an element of type P.t in a function? I tried to pass an int but it didn't work
<octachron_>
it depends on `P` signature. If you wrote something like ` module P: sig type t ... end = ... ` you have hidden the fact that `P.t` = `int`
<vsiles>
you either expose that P.t = int or provide a lift function int -> P.t
<octachron_>
then you can only generate a value of type `t` with function within `P` that generates a `t` from other (reachable) type (not necessarily `int`).
<jco>
thank you for your answers! I think I'm in your first case octachron_ because P is of signature Priority, which already defines the type t
<jco>
then I could create a lift function as vsiles suggested
<octachron_>
You can also expose the type equality with `Priority with type t = int`
ziyourenxiang has quit [Ping timeout: 268 seconds]
Haudegen has quit [Quit: Bin weg.]
kalio has quit [Ping timeout: 276 seconds]
kalio has joined #ocaml
FreeBirdLjj has quit [Remote host closed the connection]
tane has quit [Quit: Leaving]
dborisog has joined #ocaml
FreeBirdLjj has joined #ocaml
FreeBirdLjj has quit [Ping timeout: 252 seconds]
bartholin has joined #ocaml
dborisog has quit [Ping timeout: 276 seconds]
brettgilio has quit [Ping timeout: 245 seconds]
FreeBirdLjj has joined #ocaml
Haudegen has joined #ocaml
FreeBirdLjj has quit [Remote host closed the connection]
FreeBirdLjj has joined #ocaml
jco has joined #ocaml
vicfred has quit [Quit: Leaving]
<jco>
So yes, I'm still puzzled with the fact that some function taking a element t from the module P (P.t, that is) I gave earlier
<jco>
which is of type int, doesn't accept an integer
FreeBirdLjj has quit [Remote host closed the connection]
<jco>
octachron_: damn, the Priority with type t = int thing works
<jco>
thanks! but what does it do exactly?
<jco>
why it makes my function accepting P.t elements accepting 0 now?
yowera2 has joined #ocaml
<yowera2>
Heya!, Im trying to get a ocaml arm crosscompiler on x86_64, but somehow It ends up with arm binaries and tries to use them within the build. What is the current status of crosscompilation? Im using the following flags on the configure script: "--host=arm-linux-gnueabihf \--target=x86_64-pc-linux-gnu \"
sagax has quit [Ping timeout: 246 seconds]
sagax has joined #ocaml
sagax has quit [Client Quit]
<companion_cube>
fwiw i don't `open containers` either
<narimiran>
companion_cube: ok, that settles it :)