You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I was installing rust via rustup, but then it took me a while to realise that cbindgen
is needed, aka this project here.
Would it not be more convenient for just about everyone involved if cbindgen could
become official part of the rust/rustc toolchain? I think the more that is integrated,
when needed for compiling something, the better from a user's and downstream
dev perspective. At the least I would have preferred to have had cbindgen as part
of rust/rustc + rustup. It would have been easier rather than end up being frustrated
with rust not really working (because I was not aware of needing cbindgen as well),
whereas if cbindgen would be part of rust/rustc, we downstream people could
bypass one additional hurdle. Keep in mind is that we don't quite have as much
experience with regard to rust yet, but rust is also needed for other parts now
(e. g. librsvg to name but one).
The text was updated successfully, but these errors were encountered:
Would it not be more convenient for just about everyone involved if cbindgen could become official part of the rust/rustc toolchain?
What's the advantage of this? cbindgen is just tooling, and we have the [build_dependencies] section of cargo.toml to specify this is needed. The rust ecosystem is also built quite well around making tooling easy to find and use. and rust has a great deal of support for tooling.
I think the more that is integrated,
when needed for compiling something, the better from a user's and downstream
dev perspective.
I actually think that this is worse from a user's perspective. And from a dev's. If it's integrated, then all of its configuration options are now part of rustc.
I think encapsulation is a very powerful paradigm to adhere to. It keeps different things separate.
I think if you were having issues, that's probably the fault of the codebase not providing proper installation instructions, or potentially even setting it up wrong. Cbindgen can be invoked from the library's build script so that users don't need to do anything manually, not even download it.
Not sure if there is any support for this, but recently I had to compile gjs (https://www.linuxfromscratch.org/blfs/view/svn/gnome/gjs.html) and
needed firefox mozjs for that, which in turn needs cbindgen.
I was installing rust via rustup, but then it took me a while to realise that cbindgen
is needed, aka this project here.
Would it not be more convenient for just about everyone involved if cbindgen could
become official part of the rust/rustc toolchain? I think the more that is integrated,
when needed for compiling something, the better from a user's and downstream
dev perspective. At the least I would have preferred to have had cbindgen as part
of rust/rustc + rustup. It would have been easier rather than end up being frustrated
with rust not really working (because I was not aware of needing cbindgen as well),
whereas if cbindgen would be part of rust/rustc, we downstream people could
bypass one additional hurdle. Keep in mind is that we don't quite have as much
experience with regard to rust yet, but rust is also needed for other parts now
(e. g. librsvg to name but one).
The text was updated successfully, but these errors were encountered: