Re: On Boost and modules

From: Boris Kolpackov (boris_at_[hidden])
Date: 2024-04-10 04:33:31

Ruben Perez via Boost <boost_at_[hidden]> writes:

> > I don't see what's so difficult about this requirement.  Just put your
> > #includes above your imports, and don't use imports in header files.
> > Header-based code will automatically follow these rules by virtue of not
> > using imports.  Module-based code will naturally follow these rules by
> > virtue of not using header files, and by putting the #includes above the
> > imports.  Or am I missing something here?
> >
> 
> What I understood (please correct me if I'm wrong) is that importing a
> module that uses an include in its global module fragment and then
> importing another module does violate this rule.

I tend to agree with Rainer, this would only become an issue if you
start putting imports into headers (and only for modules that can
also be consumed via headers).

In particular, I believe this specific issue (ignoring declarations
that have already been imported/included) is limited to a single
translation unit. But maybe I am wrong. Can you elaborate on the
scenario that you think is problematic? In particular, I think
the "and then importing another module" part is missing the
"that does XXX" (maybe "that uses an import instead")?

I've tried the following with Clang 18 and libc++'s std module
and everything seems to work without any issues:

// a.mxx
module;
#include <iostream>
export module a;
export void a (std::ostream& os) {os << "a";}

// b.mxx
export module b;
import std;
export void b (std::ostream& os) {os << "b";}

// c.mxx
//
export module c;
import a;
import b;
import std;
export void c (std::ostream& os) {a (os); b (os);}

I've tried different orders of importation in c.mxx without
any changes in the result.