Malcolm Cohen wrote:
> Hi David,
>
> I know this is technically after the deadline, but...
>
> The IMPORT statement is ambiguous. [260:4-5] says that
> "An entity that is imported in this manner and is defined in the host
> scoping unit shall be explicitly declared prior to the interface body."
> It is not clear how (or whether) this interacts with the IMPORT with no
> list of names; this form makes everything accessible that could be, so
> does the prior specification requirement apply
> (a) to everything made accessible
> (b) to everything made accessible and actually used
> (c) not to this form at all.
> It is not clear to me (since this was a J3 mod after IMPORT was added,
> at a meeting I didn't attend) which this is meant to be. Well, I'm
> pretty certain that no-one thought about it! If they did, presumably
> they had either (a) or (c) in mind; probably (c), but that would lead
> to potential circularity and inconsistency. But (a) seems rather
> draconian.
>
> I think that we should consider adding some words to clarify this.
Why discard theory (b)? That seems to me to be the one that was probably meant.
An IMPORT without a list of names has exactly the same effect as an IMPORT that
lists those it actually uses. I think the same restrictions should apply.
Cheers,
John.
|