MaskRay wrote:
> > FWIW, we saw failures at Google (where, to the best of my knowledge, we
> > aren't using named modules at all) that look like this:
> > ```
> > error: '#include <filename>' attaches the declarations to the named module
> > '.get', which is not usually intended; consider moving that directive
> > before the module declaration [-Werror,-Winclude-angled-in-module-purview]
> > 14 | #include <stddef.h>
> > ```
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> > So there's probably some problems with this patch? is this enough to go on,
> > or would a reduced test case be required to address the issue? (reducing
> > modules issues is a bit difficult/expensive, or I'd have provided it up
> > front)
>
> Since the patch itself is pretty simple, if there is a problem, it should
> come from the implementation `Preprocessor::isInNamedModule()`. It should be
> helpful to provide a reproducer so that we can be sure we're facing the same
> issue.
See #71134 :)
```cpp
struct module {}; void foo(module a);
#include <stdio.h>
```
https://github.com/llvm/llvm-project/pull/69555
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
