In my journey to learning C, I’ve come across header files, which are used to (I’m assuming) define a prototype for the source file, as well as structure modules. This feature, in my opinion, is pointlessly not just redundant, but possibly a source for pitfall. The same information can probably be extracted from the source code, if not for the restrictions of the language specification in C.
Say, if I have a GTK project, I will have to use the preprocessor directive, that will require the use of GTK headers that look something like , and they’re usually in the system path. How do modern languages, like Rust, Zig or Go deal with this situation, where shared libraries are used?
I’ve never used Rust or Zig, but for Go: (disclaimer: this is all from memory, so there may be inaccuracies or out-of-date information here)
Go does not allow circular references between modules. That restriction allows the compiler, when compiling a module, to not only put the compiled machine code in the resulting object file for that module but also the information that in C would be required to be in a header file (i.e. type definitions, function signatures, and even complete functions if they’re considered candidates for inlining, etc.). When compiling a module that imports others, the compiler reads that stuff back out of those files. Essentially a compiled Go library has it’s auto-generated “header file” baked-in.
In older versions this was actually human-readable: an early part of the object file would essentially look like trimmed-down Go when opened in a text editor. IIRC they’ve switched to a binary serialization format for this some years back, but AFAIK it still essentially works the same.
I guess when comparing to C or C++, you could compare this to automatically generating pre-compiled headers for every module, except the headers themselves are also auto-generated (as you alluded to in your post).
If by “shared library” you mean a dynamically linked one: IIRC Go does allow shared libraries to be used, but by default all Go code is linked statically (though libraries written in other languages may be dynamically linked by default, if you import a module that requires it).
This is the same for Rust.
go has an abi, if unstable between versions.
https://pkg.go.dev/internal/abi
rust foundation has stated that they will never have an abi.
I think both and zig support C ABI. Because Rust has no internal abi, projects will always be structurally incompatible with dynamic loading. that leaves rust with FFI.
Ah yeah, I guess, my interpretation of that quote wasn’t quite right.
You can build a shared library in Rust, but it will need to be called via the C ABI or the WebAssembly ABI.
It is also possible to call a C ABI library in Rust (and virtually any other language), as well as a WebAssembly ABI library.
So, technically you can do shared libraries that way, but because the C ABI and WebAssembly ABI are significantly more limited compared to what you want to be passing around internally in Rust, you’ll only really want to use these in special cases.
Rust doesn’t particularly like dynamic libs. It compiles libraries you use on your dev machine, so it knows how you’ll use the libraries, which allows it to make various assumptions and optimizations.
For example, Rust can do generics via monomorphization (as opposed to vtables), which means it will generate the generic library code for each type that’s passed into the generic type argument. This is useful, because it can fully optimize that code, but also because it retains type information, despite the use of generics.