<recherche>
trying to compile ocamldebug. when i do `make && ocamlrun debugger/ocamldebug hello.bc`, i get `Sys_error("/usr/local/lib/ocaml/compiler-libs/topdirs.cmi: No such file or directory")`. any ideas?
gravicappa has quit [Ping timeout: 260 seconds]
<octachron>
You need to include the in-tree standard library with "-I stdlib"
bartholin has joined #ocaml
gravicappa has joined #ocaml
mro has joined #ocaml
<Armael>
dune question: I have a test binary that produces several files (that I want to commit in git and use as expect files); right now these files get produced in _build/default/test/
<Armael>
and I don't know how to tell dune to copy those files outside of _build, except by listing them explicitly (which is annoying)
bobo has joined #ocaml
<Armael>
I guess I could make the entire output of the binary to be in one file, but then `git diff` becomes less useful
spip has quit [Ping timeout: 268 seconds]
<Leonidas>
Armael: Is there a way to hook into `dune promote`?
Techcable_ has joined #ocaml
<Leonidas>
Since that seems to be the mechanism that you want but I don't know if users in dune can build their own promotion rules
Techcable has quit [Ping timeout: 265 seconds]
<Armael>
I don't mind doing the promotion by hand using git
<Armael>
(also I don't know if it's possible to hook into dune promote)
<Armael>
the issue seems to be more: how do I make dune aware that these files were produced
spip has joined #ocaml
bobo has quit [Ping timeout: 268 seconds]
Enjolras_ is now known as Enjolras
wonko has joined #ocaml
haesbaert has quit [Read error: Connection reset by peer]
average has quit [Quit: Connection closed for inactivity]
haesbaert has joined #ocaml
mro has quit [Read error: Connection reset by peer]
mro_ has joined #ocaml
<recherche>
re: ocamldebug, it was -topdirs-path toplevel that i needed. what's the difference between this and -I?
<d_bot>
<Président daren le pétrolier> hi, do you know the reason why you can't partially apply a constructor in OCaml ?
<d_bot>
<Président daren le pétrolier> to make things like ``"bar" |> Foo``
<theblatte>
constructors are not functions so you cannot use them as you would functions
<d_bot>
<octachron> In particular, constructors are guaranteed to be pure and they can have an arity different from one.
<d_bot>
<Président daren le pétrolier> thanks for your answers, so it's more a design choice than a performance concern ?
mro has quit [Ping timeout: 268 seconds]
<d_bot>
<octachron> It is a design choice to have syntax reflects performance concerns and a constraint imposed by the value restriction which implies that `cons x xs` and `x::xs` may not have the same type.
<d_bot>
<Président daren le pétrolier> okay, thanks a lot
<vsiles>
is there a way to list all the dune lang version supported by dune ?
<d_bot>
<andreypopp> does it mean List.map won't overflow the stack anymore?
<d_bot>
<andreypopp> huge
<olle>
yuge
<olle>
or just do for-loops? :D
<olle>
"NEVER! We need more science so we don't have to write for-loops."
mro has joined #ocaml
waleee has joined #ocaml
mro has quit [Remote host closed the connection]
<companion_cube>
or use an existing tailrec version of List.map
<companion_cube>
but this makes it far easier to make one
mro has joined #ocaml
recherche has joined #ocaml
<theblatte>
companion_cube: oooh, yummy!
<companion_cube>
right? although it's opt-in
<companion_cube>
but that might be for the best
<octachron>
or more precisely to match reasonable code with reasonable performance.
<octachron>
and for stack-overflow, it seems that we will have also have increased default stack size in 4.14.
<companion_cube>
ohhhhhh yes
<companion_cube>
can it be adjusted dynamically, too?
mro has quit [Ping timeout: 268 seconds]
<companion_cube>
for some programs it makes sense to use very large stacks
<companion_cube>
cause making things tailrec is infeasible
<octachron>
the stack size will still be adjustable dynamically from the OS
<octachron>
OCaml executable will try to increase the stack size to min 1GB hard_OS_limit if the current value in the environment is the default value.
<companion_cube>
awesome
<companion_cube>
zozozo: you hear that? 😁
wonko has quit [Ping timeout: 268 seconds]
mro has joined #ocaml
<zozozo>
companion_cube: yup saw the pr for the stack space
<zozozo>
it's really nice (though you still may need ulimit -s unlimited, or to wait for multicore to really make stack overflows a thing of the past)
<d_bot>
<mbacarella> is there an example minimum github action for ocaml that will test-build your repo on each push?
<d_bot>
<mbacarella> starting from zero knowledge on how to set up CI on github?
<companion_cube>
zozozo: I think I'm fine if it means I can't process terms of depth > 1M or something like that :D
<zozozo>
companion_cube: that might be fine for most use-cases indeed, ^^
<companion_cube>
even yours, no? :p
<companion_cube>
1GB of stack seems… plenty
<zozozo>
well...
<zozozo>
i'd have to try, but it wouldn't surprise me if there wre some instances of smtlib problems which are, pathological (e.g. a sequence of around 1M let-bindings, or more...)
mro has quit [Remote host closed the connection]
<companion_cube>
I'd be surprised
<companion_cube>
1M let bindings still means 1M _distinct_ terms
<companion_cube>
that seems like a lot, even for a SMT solver :)
<d_bot>
<mbacarella> yes i was. it linked me to that as a minimum test case 🙂
mro has joined #ocaml
mro has quit [Ping timeout: 268 seconds]
waleee has quit [Quit: WeeChat 3.3]
mro has joined #ocaml
favonia has joined #ocaml
recherche has quit [Quit: Client closed]
hendursa1 has quit [Quit: hendursa1]
hendursaga has joined #ocaml
Haudegen has quit [Quit: Bin weg.]
bartholin has quit [Quit: Leaving]
average has joined #ocaml
olle has quit [Ping timeout: 260 seconds]
mbuf has quit [Quit: Leaving]
mro has quit [Quit: Leaving...]
<d_bot>
<darrenldl> is m4 good for code gen?
<companion_cube>
afaict m4 is good for absolutely nothing
<d_bot>
<darrenldl> oh
<d_bot>
<darrenldl> : D
<d_bot>
<darrenldl> a project of similar scale/scope is relying on m4 for a lot of macro ish stuff, and i was wondering why havent they gone with writing code gen in haskell (they built the prover in haskell) but in m4 instead
rgrinberg has joined #ocaml
<d_bot>
<mseri> I think there are premade examples in the setup-ocaml repo
<d_bot>
<mseri> You can make it much shorter than what you see there
<d_bot>
<mseri> Oh yes, what @undu said
<companion_cube>
@darrendl your project is a prover too?
Haudegen has joined #ocaml
wonko has joined #ocaml
vicfred has joined #ocaml
<d_bot>
<mbacarella> I have somehow consumed 1250 github compute minutes this morning build testing my repo against mac/ubuntu/windows. that's oddly satisfying
waleee has joined #ocaml
average has quit [Quit: Connection closed for inactivity]
rgrinberg has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
favonia has quit [Remote host closed the connection]
favonia has joined #ocaml
Serpent7776 has quit [Read error: Connection reset by peer]
Serpent7776 has joined #ocaml
<d_bot>
<Continuation Calculus> how can i do it so that i don't get diffs logs on my builds when using deriving_inline?