Language Change Specification for All Interface Lists Can Be Ordered Proposal

Voting Results: Cast your votes here

Yes:

1. Kevin Jennings - 2016-11-07
2. Patrick Lehmann - 2016-11-19 - ver 2
3. Thomas Preusser - 2016-11-27
4. Morten Zilmer - 2016-12-11
5. Martin Thompson - 2016-12-12
6. Ryan Hinton - 2016-12-19
7. Martin Zabel - 2017-01-19
8. Lieven Lemiengre - 2017-01-27
9. Jim Lewis - 2017-02-16 - ver 3

No:

Abstain:

1. Brent Hayhoe - 2017-02-16 Version 3 - Abstain due to lack of personal time for review.

Revision Notes

• Adds edits to the examples
• Restored red text

Comments

The link to the source document seems to be broken. -- Main.PatrickLehmann - 2016-11-19

Fixed. -- Main.KevinJennings - 2016-11-21

The current wording is too vague and risks mutually inconsistent implementations. As it is, it would allow arbitrary cross referencing. However, it should either only allow back references to previously declared names or explicitly forbid cyclic references. -- Main.ThomasPreusser - 2016-11-21

How about adding "as long as the name in the interface declaration appears after the name in the interface definition" to the end of the sentence? Am open to other suggested wording as well. -- Main.KevinJennings - 2016-11-21

Going for back references only, I would suggest: A name that denotes an interface declaration in a generic interface list or a port interface list or a parameter interface list may appear in a subsequent interface declaration within the interface list containing the denoted interface declaration. -- Main.ThomasPreusser - 2016-11-21

Agreed and updated in Version 2. -- Main.KevinJennings - 2016-11-21

Can we shorten the paragraph edited to:

A name that denotes an interface declaration in a generic an interface list may appear in an a subsequent interface declaration within the same interface list%RED. containing the denoted interface declaration.

-- Jim Lewis - 2017-02-16

I am just wondering what ruling defines whether Y1'subtype refers to the formal or rather the actual. This seems not to be clear at least for scalars.

-- Thomas Preusser - 2017-02-16