In the email you partially quoted, I went on to give examples of what I meant by components:

> The key issue is not, IMO, how you control a “checking mode” that applies to a TU or the program, the issue is how you apply various “checking modes” to different components (e.g., libraries).

I think that combined with the general description should illustrate the problem.

Also, if you go to the clang link in the paper and look at the clang approach to a label-like mechanism, it should make it even clearer.

John.

> On Sep 30, 2025, at 1:44 AM, Timur Doumler via SG21 <sg21_at_[hidden]> wrote:
>
> Hi John,
>
>> On 29 Sep 2025, at 23:14, John Spicer via SG21 <sg21_at_[hidden]> wrote:
>> But that misses the essential point of p3835r0 that the semantics need to apply to software “components” and not to translation units or to the entire program.
>
> What are "components", exactly? I am not aware of any such entity being specified in the C++ Standard or existing in the C++ ecosystem today. Neither am I aware of any mechanism to determine whether an inline function belongs to such a "component". Could you please clarify what your expectation is here? Otherwise it might be hard to understand the essential point of p3835r0 that you are trying to bring across.
>
> Thanks,
> Timur
>
> _______________________________________________
> SG21 mailing list
> SG21_at_[hidden]
> Subscription: https://lists.isocpp.org/mailman/listinfo.cgi/sg21
> Link to this post: http://lists.isocpp.org/sg21/2025/09/11204.php