Re: [sip-clf] Request for Consensus: Adopt ASCII

[Peter Musgrave <[e-mail]>]

Hi Vijay, 

That works for me.

Peter

On 2010-12-03, at 9:48 AM, Vijay K. Gurbani wrote:

> On 12/03/2010 07:07 AM, Peter Musgrave wrote:
>> Hi Vijay,
>> 
>> Maybe we can do something simple (but not perfect)
>> - for empty field
>> ? for unparsable field
>> 
>> In practice a tag or other field which contains the literal text - or
>> ? would not be distinguishable from empty/unparsable, but I view that
>> as a very remote case. We could go down the rat-hole of escaping
>> them, but then e.g. a tag could legitimately contain the exact
>> escaped sequence etc. etc.
> 
> Peter: I agree that a tag field consisting of only one character --- an
> unadorned "-" --- and the Call-ID field consisting of only one character
> --- an unadorned "?" --- will be extremely rare in practice (although
> ABNF allows them.)
> 
> If these characters are part of a sequence of other characters, then
> of course, there is no ambiguity.  Ambiguity arises if they are the
> only character for that field, and in such a case, we could say
> something like the following:
> 
>  If the To- or From- tag fields contain only one character, and that
>  character is the literal "-", the implementation SHOULD insert an
>  escaped %2D in the SIPCLF record.  Similarly, if the Call-ID field
>  contains only one character, and that character is the literal "?",
>  the implementation SHOULD insert an escaped %3F in the SIPCLF record.
> 
> This way, the SIPCLF log record can still contain "-" and "?" and if
> this is the only character in the field, then the reader can assert
> that the fields were empty and unparseable, respectively.
> 
> Thanks,
> 
> - vijay
> -- 
> Vijay K. Gurbani, Bell Laboratories, Alcatel-Lucent
> 1960 Lucent Lane, Rm. 9C-533, Naperville, Illinois 60566 (USA)
> Email: vkg@{alcatel-lucent.com,bell-labs.com,acm.org}
> Web:   http://ect.bell-labs.com/who/vkg/