Re: [dnsext] WGLC: RFC6195bis IANA guidance

[Donald Eastlake <[e-mail]>]

Hi Chris,

On Fri, Jul 6, 2012 at 2:56 PM, Chris Thompson <[e-mail]> wrote:
> On Jul 5 2012, Donald Eastlake wrote:
>
>> Hi Dick,
>>
>> On Thu, Jul 5, 2012 at 6:06 PM, Dick Franks <[e-mail]> wrote:
>>>
>>> Donald,
>>>
>>> My apologies for late response.
>>>
>>>
>>> On 22 June 2012 19:10, Donald Eastlake <[e-mail]> wrote:
>>> ...
>>>>>
>>>>> [3.1 paragraph 4]
>>>>> and [3.2 paragraph 4]
>>>>>
>>>>> Regexes:
>>>>>
>>>>>                         [A-Z][A-Z0-9\-]*[A-Z0-9]
>>>>>
>>>>>                        (TYPE|CLASS)(0|[1-9][0-9]*)
>>>>>
>>>>> could be simplified to:
>>>>>
>>>>>                         [A-Z][A-Z0-9]*
>>>>>
>>>>>                        (TYPE|CLASS)[0-9]*
>>>>
>>>>
>>>> That's not simplification, that's change.
>>>
>>> A simplification of the underlying production rule for RRTYPE mnemonics,
>>> which inevitably flows through to the regex.
>>
>>
>> OK, it is a simplification of the RegEx, but one I don't agree with.
>
>
> Regardless of the hyphen question in the first regexp, I think the
> change to the second is probably desirable. I don't think RFC 3597
> makes it totally unambiguous that unnecessary leading zeros are not
> allowed after TYPE or CLASS, and it is sensible to protect parsers
> that are sloppy in this respect.
>
> Otherwise, one could argue that TYPE65536 ought to be allowed for
> a new RR type mnemonic, because it clearly isn't meaningful as a
> generic one.

I agree and that change is actually already in the -03 version of the draft.

> This is all a bit reminiscent of the deprecation of all-digit TLDs.
> Ought ".000" or ".256" to be allowed because they are would not
> be part of the natural representation of IPv4 addresses?

Not that relevant to our current discussion but it is my personal
opinion that all-numeric TLDs should be prohibited regardless of
length.

Thanks,
Donald
=============================
 Donald E. Eastlake 3rd   [phone] (cell)
 155 Beaver Street, Milford, MA 01757 USA
 [e-mail]

> --
> Chris Thompson               University of Cambridge Computing Service,
> Email: [e-mail]    New Museums Site, Cambridge CB2 3QH,
> Phone: [phone]       United Kingdom.