Re: The use of "name-list" in the core IDs.

[nisse%lysator.liu.se@localhost (Niels Möller)]

"denis bider" <ietf-ssh%denisbider.com@localhost> writes:

> I believe the traditional syntax for name-lists might also be exampled ("",
> "zlib", "zlib,none"), because the ("zlib", "none") notation is not what
> actually appears in the protocol.

I think it was I who wrote the original version of this text. The
point is to make it clear that the three given examples represents
lists containing zero, one and two elements, on a somewhat abstract
level. In particular, extra clarity is needed for the case of the
empty list.

To me, the proposed text is clear enough.

:        Examples:
: 
:        value                      representation (hex)
:        -----                      --------------------
:        (), the empty name-list    00 00 00 00
:        ("zlib")                   00 00 00 04 7a 6c 69 62
:        ("zlib", "none")           00 00 00 09 7a 6c 69 62 2c 6e 6f 6e 65


> Also I'm not sure how the charset is restricted, but I think it's
> restricted at least to US-ASCII, if not also the printable subset of
> it.

I think "anything but comma and NUL" is the only appropriate
restriction at this place (I would even allow NUL, but I guess using
NUL in the elements might be confusing). Context implies additional
requirements: For lists of algorithm identifiers, there are
restrictions to ascii-only, on the length of the elements, and on the
use of the @-character. For lists of language tags, requirements are
different (I'm not really familiar with them). I don't think all those
details need or should be duplicated in the description of the
name-list type. We may want to reuse the same type for new purposes,
even ones using non-ascii identifiers in some way.

> Otherwise I believe it's fine.

To me the proposed text looks fine as is. I won't object if anybody
wants to tweak the syntax used in the examples, but I don't think it
is needed.

Regards,
/Niels