On Thu, Sep 1, 2011 at 6:36 AM, Bill Roome <[email protected]> wrote:
> There's also the question of what characters are legal in a map-vtag.
>
> {7.5.1} says PID names have to be US-ASCII, alphanumeric, 64 characters
> max. Why not do the same for map-vtags?

Well, so 7.5.1 also states that they must be alphanumeric characters
or the '.' character. Given that they are opaque and
not-to-be-standardized identifiers, then we could probably be much
less restrictive here.

How about for map-vtags: Any US-ASCII character within 0x20 and 0x7E
(before encoding according to JSON, of course).

Also, it has been requested that PID names also have a larger
character set.  Would there be any complaints with doing the same for
PID names (with the existing rule still in place about '.')?  The only
(admittedly-small) concern I would have is that it might be confusing
from a human-readability perspective if we ever did something with
hierarchical PIDs and other punctuation characters were combined with
the '.' in the same PID name.

Thoughts?
Rich

>
>        - Bill Roome
>
>
>>Date: Sat, 27 Aug 2011 07:58:47 +0200
>>From: Enrico Marocco <[email protected]>
>>To: "[email protected]" <[email protected]>
>>
>>Server implementations seen at the interoperability event have shown a
>>broad variety of formats for the 'map-vtag' field, with lengths varying
>>from 1 char to tens of chars. It has been pointed out that some explicit
>>indication in the specs about the length and perhaps the format of the
>>field could lead to a more homogeneous use of it.
>>
>>Possible options (please express and possibly motivate your preference):
>>
>>  1. leave the format and length unspecified, possibly allowing
>>     implementations to override its semantics (e.g. piggybacking
>>     additional/proprietary information);
>>
>>  2. define an upper limit to the length of the field, but leave format
>>     unspecified;
>>
>>  3. other.
>>
>>--
>>Ciao,
>>Enrico
>
>
> _______________________________________________
> alto mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/alto
>
_______________________________________________
alto mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/alto

Reply via email to