<discocaml>
<darrenldl> what's a good choice for fast and stable (at least more stable than marshal) binary serialisation these days in ocaml land?
<discocaml>
<darrenldl> primarily want very fast unpacking
neuroevolutus has joined #ocaml
<discocaml>
<yawaramin> anything with a schema should unpack pretty quickly, right? because it won't have to read any field names
neuroevolutus has quit [Ping timeout: 256 seconds]
neuroevolutus has joined #ocaml
<discocaml>
<darrenldl> yeah true, but i want to be lazy and pick something well supported/finished
<discocaml>
<darrenldl> ugh ocamlwiki dot com is flooding my search results again in duckduckgo. anyway.
euphores has quit [Quit: Leaving.]
euphores has joined #ocaml
bartholin has joined #ocaml
neuroevolutus has quit [Ping timeout: 256 seconds]
<discocaml>
<timmcgilchrist> Protobufs has been decent so far. There’s two main libraries for it in OCaml. I’m using it for serializing program backtraces and logging data.
<discocaml>
<0x28west> i'm getting this error when i run `opam init -y` on window i have visual studio build tools 2019/2022 installed with all sdk's installed as well
<discocaml>
<yawaramin> i recommend asking in #windows-support and giving details of exactly what you did. eg, open PowerShell, run commands A, B, C, following instructions on XYZ page. and so on
torretto_ has joined #ocaml
torretto has quit [Ping timeout: 260 seconds]
euphores has quit [Ping timeout: 252 seconds]
<discocaml>
<getzapped.> Do recursive modules mess with type visibility? I have a module Env with a type t, and another functor that takes a module of Env's signature, and so a function in that functor takes an Env.t. Env is recursive, so I pass itself to the functor to create a module visible within Env. When I pass a value of type t from Env to the function taking Env.t, the compiler complains about the value being of `type t = <actual type>, but a value was expe
<discocaml>
<getzapped.> Do recursive modules mess with type visibility? I have a module Env with a type t, and another functor that takes a module of Env's signature, and so a function in that functor takes an Env.t. Env is recursive, so I pass itself to the functor to create a module visible within Env. When I pass a value of type t from Env to the function taking Env.t, the compiler complains about the value being of `type t = <actual type of Env.t>, but a value
<discocaml>
<getzapped.> fixed it by the functor taking the actual type instead of the abstract Env.t, and having identity converting functions in Env, but it doesnt make sense to me
Anarchos has joined #ocaml
Absalom1 has joined #ocaml
kurfen_ has joined #ocaml
gentauro_ has joined #ocaml
adrien_ has joined #ocaml
neuroevolutus has joined #ocaml
bartholin has quit [*.net *.split]
toastal has quit [*.net *.split]
adrien has quit [*.net *.split]
kurfen has quit [*.net *.split]
gentauro has quit [*.net *.split]
Absalom has quit [*.net *.split]
Absalom1 is now known as Absalom
neuroevolutus has quit [Ping timeout: 256 seconds]
<discocaml>
<sim642> I can't exactly imagine what you have and where the problem appears, but I think recursive module definitions do have some limitations. Like type equalities not being available in the middle of recursive module definition, but only after it or something. This might've come up recently on Discuss as well
<discocaml>
<getzapped.> Thanks for the info. I ended up ditching the functor because I remembered the and keyword existed. The functor became useless when I made them properly recursive, then the compiler led me to the manual page for safe modules
toastal has left #ocaml [Disconnected: Hibernating too long]
myrkraverk_ has quit [Read error: Connection reset by peer]
myrkraverk_ has joined #ocaml
spew has joined #ocaml
spew has quit [Remote host closed the connection]
spew has joined #ocaml
Anarchos has quit [Quit: Vision[]: i've been blurred!]