Mismatch in size of std::string

[dtolnay]

On my machine this input header results in the following generated Rust struct, which looks reasonable, except the Rust side and C++ side disagree about the size of std::string.

#pragma once
#include <cstdint>
#include <string>
struct S {
  explicit S(uint32_t i);
  std::string s;
  uint32_t i;
};

// generated by autocxx:

#[repr(C)]
pub struct S {
    pub s: ::cxx::CxxString,
    pub __bindgen_padding_0: [u32; 6usize],
    pub i: u32,
}

I'll share a PR to help reproduce. Possibly bindgen run by autocxx is finding a different set of system headers than the C++ build?

Here is the debug printing I used to confirm the issue:

S::S(uint32_t i) : i(i) {
  std::cout << "C++:" << (size_t(&this->i) - size_t(this)) << std::endl;
}

let s = ffi::cxxbridge::S_make_unique(1);
println!("Rust:{}", (&s.i as *const _ as usize) - (&*s as *const _ as usize));
println!("{}", s.i);

Expected output is that C++ and Rust print the same size and the third line is 1, but the actual output is:

C++:32
Rust:24
0

[adetaylor]

Per comments on #18 this is intentional because of this prelude without which bindgen can't cope with std::string at all (and, as you hint, different STLs could theoretically give different implementations so we probably can't make better assumptions - I must admit I have not checked the C++ standards though). I am going to treat this as a duplicate of #18, which remains in brief "field access to non-POD types is doomed"!!

[dtolnay]

Ah that's too bad. :( It would have been really slick to be able to support this without accessor methods. I have some ideas for how to make it work anyway without accessors and without requiring to know the field sizes at codegen time, which I will write up.